Community Research and Development Information Service - CORDIS

FP7

CONCORDE Report Summary

Project ID: 620130
Funded under: FP7-JTI
Country: Spain

Final Report Summary - CONCORDE (Flight Operations for Novel COntinuous DEscent - cONCOrDE)

Executive Summary:
CONCORDE (Flight Operations for Novel Continuous Descent) is a Research & Development project funded by the Clean Sky Joint Technology Initiative and project-managed by Pildo Labs. It is designed to evaluate the viability of the Time and Energy Managed Operations (TEMO) concept.

TEMO is a green concept that enables aircraft to fly optimum speed trajectories whilst adhering to time constraints for spacing. For this purpose, TEMO algorithm has been developed using non-linear programming (NLP) solver for the trajectory prediction and optimisation. Once the optimum trajectory is computed, TEMO use of the aircraft autopilot to guide the aircraft using Speed-on-Elevator (SOE) control and with thrust set to idle. In case that TEMO cannot plan an energy neutral trajectory, the concept makes use of the autothrottle and auto speedbrakes systems to add or remove energy from the system at certain points of procedure.

Throughout the CONCORDE project framework, TEMO concept has been successfully evaluated by defining, executing and analysing different simulator and real experimental flight campaigns in order to demonstrate its operational feasibility and environmental benefits.

Initially, the Experiment Plan was drafted detailing the approach for testing the TEMO concept in two different simulators. The main points defined per each experiment were: analysis of TEMO software version to be used, research questions and hypotheses to be taken into consideration when analysing the experiment results, validation objectives and acceptance criteria to assess the experiment and simulator experiment scenarios.

According to the experiment plan, both TEMO simulation experiment were executed and analysed. First flight simulator experiment was performed at topic manager Amsterdam premises along 3 days (July 8th, 9th and 10th 2014) with the aim of demonstrating that TEMO concept reaches a Technical Readiness Level (TRL) 5, in which technology is validated in a relevant (simulated) environment. Second simulator experiment was performed at topic manager Braunschweig premises along 3 days (September 22nd, 23rd and 24th of 2014) with the aim of investigating the possibility of using TEMO on a modern aircraft with unmodified or only slightly modified avionic systems.

Afterwards, and based on the good results obtained in the simulator experiments, in October 2015 topic manager executed a flight trial campaign using a Cessna Citation II research aircraft at Eelde airport in the Netherlands. The trials, executed along 6 days, consisted of several descents and approaches demonstrating the TEMO feasibility and environmental benefits. CONCORDE consortium supported these experiment by providing technical assistance and contributing to the assessment tasks. Different TEMO conceptual variants were flown and it was demonstrated that the TEMO concept is feasible and enables arrival with timing errors below 10 seconds.

Finally, the different aspects involving TEMO concept integration into a commercial aircraft were analysed by identifying which should be the new elements on-board, TEMO-avionics interfaces and certification difficulties that could be faced in the process.

Project Context and Objectives:
Continuous Descent Operations
-----------------------------------------------
According to ICAO, a Continuous Descent Operation is an operation, enabled by airspace design, procedure design and ATC facilitation, in which an arriving aircraft descends continuously, to the greatest extent possible, by employing minimum engine thrust, ideally in a low drag configuration, prior to the final approach fix.

CDO is one of several tools available to aircraft operators and ANSPs to increase flight predictability and airspace capacity, while reducing noise, fuel burn, emissions and controller-pilot communications. Over the years, different route models have been developed to facilitate CDO, and several attempts have been made to strike a balance between ideal fuel-efficient and environmentally friendly procedures and the capacity requirements of a specific airport or airspace.

However, the reality of the extant situation throughout the majority of the European network is that airspace congestion and capacity issues lead to operational limitations that have a negative impact on CDO utilisation. Thus, at present, CDO facilitation is usually confined to periods (or areas/airports) of low traffic density and generally from lower levels. ATC instructions often lead to trajectory deviations which result in non-optimal operations.

Time and Energy Managed Operations
---------------------------------------------------------
With the aim of improving the CDOs concept, a Trajectory Optimization Technology has been developed that optimizes the descent profile by utilizing (closed loop) control of both altitude and speed, to meet an absolute or relative time constraint. The concept behind this technology is called TEMO (Time and Energy Managed Operation).

TEMO is an on-board trajectory planning and guidance algorithm that optimizes the vertical approach trajectory, while satisfying required time of arrival (RTA) constraints at certain points. Thus, ATC will command these RTAs to the arriving aircraft in order to maintain separation without the need to interfere to the lateral or vertical execution of the descents. It will be the responsibility of the aircraft to compute the best trajectory that fulfils the imposed RTAs. TEMO tries to generate an airspeed-profile, which can be flown using idle thrust and no speed brake usage as a function of the aircraft characteristics (weight and aerodynamics). If such a trajectory cannot be found because a too restrictive RTA, the algorithm will look for a sub-optimal trajectory while minimizing the usage of engine throttle and speed brakes.

TEMO principle is to use strategic re-planning of trajectories in order to meet all the trajectory constraints. Therefore, TEMO is an “open-loop” solution since no immediate action to resolve time or energy deviations due to uncertainties is taken by means of any feedback or retro alimented loop. Nevertheless, the trajectory is monitored and when the error in time and/or energy exceeds certain allowable error margins, a new trajectory optimization is triggered based upon the actual flight state conditions. This is known as the trajectory re-planning process.

This re-planning algorithm is formalized as an optimal control problem, consisting in finding the best control functions (like for instance the engine throttle, lift coefficient, speed brakes settings, etc.) for a given time interval that minimize a given cost functional (like for instance fuel usage, controls usage, noise footprint, etc.). This optimal control problem is highly nonlinear and it is impossible to obtain analytic solutions. Thus, numerical methods are required to solve it that require the discretization of the problem and the approximation of the differential equations that describe the dynamics of the aircraft into a finite set of algebraic equations.

Some limitations arise, however, when this kind of problems are solved numerically. Basically, the algorithms lack from determinism since NLP solvers cannot guarantee to always find a feasible solution. Moreover, the time to solve a given problem can significantly vary as a function of the input conditions and in some runs, can be unacceptable high. Thus, if it is foreseen to implement this kind of algorithms in a real cockpit environment, operational and certification issues will arise. In this context, the implemented system will have to provide, for instance backup trajectory solutions in case the optimal control algorithm does not converge in the required time.

TEMO Previous Studies
------------------------------------
Initial batch studies of TEMO concept were carried out to test the feasibility of the concept. Subsequently a Human in the Loop study was performed to look at Human Factors aspect which helped in reaching the technology readiness level 4 (TRL-4). Yet, the model contained several important approximations, being winds calm and international standard atmosphere (ISA) assumptions perhaps the most relevant ones. The FASTOP (Fast Optimizer for Continuous Descent Approaches) project, funded by the CleanSky Joint Undertaking initiative, enhanced the current version of the TEMO algorithm in order to test it in more realistic environments, aiming at the TRL-5 gate.

CONCORDE Objectives
-------------------------------------
The CONCORDE project will define, prepare, perform and analyse two flight simulator experiments and a campaign of flight tests with the objective to test and validate the TEMO concept of operations including operational, safety and certification requirements, to identify issues in these fields and to produce plans on how mitigating them. The outcome of CONCORDE simulations aims at proving that TEMO concept has successfully reached TRL-5, both technically through simulation as well as through support documents.

More precisely, the aim of the first simulation experiment is to demonstrate that TEMO concept reaches a TRL-5, in which the technology is validated in a relevant (simulated) environment. On the other hand, the second simulation experiment pretends to investigate the possibility of using the TEMO trajectory prediction on a modern aircraft with unmodified or only slightly modified avionic system. Finally, the flight trial experiment is focused to demonstrate the TEMO concept in a relevant (real-world) environment, providing support the TEMO concept beyond TRL-5 towards TRL-6.
Project Results:
FIRST TEMO SIMULATION EXPERIMENT
----------------------------------------------------------------
TEMO flight simulations were performed at topic manager Amsterdam premises along 3 days: 7th, 8th and 9th of July 2014. The experiment was focused to demonstrate that TEMO concept reaches a Technical Readiness Level (TRL) 5, in which the technology is validated in a relevant (simulated) environment.

This was achieved by evaluating the pilot acceptance of the TEMO concept in various real scenarios while using different modes of operation: RTAs, Interval Spacing and/or Tactical Guidance.

* Participant remarks

Below, the remarks that the participants in the simulations made are summarized and grouped into different topics. These include opinions from all pilots involved, the instructor pilot and involved engineers. The opinions were gathered from group debriefings and discussions, notes taken during the simulations and remarks that some pilots made directly in the questionnaires.

•TEMO CONOPS and economic feasibility:
o All pilots involved in the simulations congratulated the TEMO team, acknowledging TEMO can be very useful and the system tested a promising functionality that could equip future aircraft.
o All agree that, at present, even small savings in fuel (and noise in some airports) are significant enough to justify the certification costs for such new technology.
o All pilots coincide in pointing out that TEMO guidance performance, in terms of fulfilling RTA/RTI, is the best they have ever seen, since state-of-the art FMS are still very sensitive to uncertainties during the descent, leading to big errors for RTA at fixes such as the IAF or the runway threshold.
o All pilots appreciate the dynamism of TEMO re-planning functionality and really felt that the trajectory was being updated continuously during the descent in order to cope with weather incertitude or ATC instructions. Some of them pointed out that state-of-the-art FMS freeze the descent plan well before the top of descent and it is not updated anymore, leading to significant time errors at metering fixes (since guidance laws are designed to keep the planned path).
o The instructor pilot remarked that all pilots were rapidly familiarised with the concept and felt comfortable flying the new procedures. He was convinced that TEMO could reduce crew workload, although some particular pilots felt that workload reduction would not be significant.
o Besides fuel and noise, the gains in capacity and in reducing ATC workload are also acknowledged. All agree that TEMO can really help in maintain or even increase capacity in dense and complex TMAs.
o Everyone agreed that TEMO operations must be coordinated with new AMAN (arrival manager) and DMAN (departure manager) research in the context of SESAR, in order to deliver correct RTAs to the different aircraft. Discussions related to the implication and responsibilities of the ATC were brought up in several occasions. Although it is well understood that ATM is out of the scope of CleanSky projects, the active participation of ATC was considered of vital importance in the future developments of the TEMO concept.
o For all experiments the RTA at the IAF was always fulfilled within the allowed time error tolerance, even in the worst-case scenario with 9kt/6o of wind prediction errors. RTA adherence at the FAP, however, was much more difficult to achieve, leading to several re-plans in some cases (too many for some pilots) or to situations where it was impossible for the system to fulfil the RTA (unable plan). At lower speeds, wind affects much more the trajectory of the airplane and therefore, time (and energy) deviations grow quicker.
o Some pilots felt that TEMO algorithm was too sensitive to weather uncertainties and were curious to know whether TEMO algorithm could handle trajectories in presence of thunderstorms or severe weather with more than 30kt of wind forecast errors (as it happens sometimes in real operations).

•Operational feasibility and potential safety issues:
o Pilots are in general satisfied with the TEMO concept and remark that after few runs it is very easy to become familiar with the system. A major concern for most pilots was the cabin layout used for the experiments, since it was a mix of systems and panels of different aircraft models. All pilots consider this fact may have induced some small errors and perhaps perturbed the results of this experiment. Some pilots asked for additional training related to the energy concept and the TEMO specific procedures in general.
o Some pilots were not used to capture the glideslope from above and all of them were almost not familiar with speed-on-elevator guidance in the flight director. In this context they claimed more training to get used to such operations (see training related with question Q4 from the post questionnaire in Appendix B). Some also asked to modify TEMO planning algorithm in order to prevent trajectories to end above the glideslope at the FAP, or perhaps to disconnect TEMO re-planning when approaching the FAP.
o Pilots, instructor pilot and engineers largely discussed the consequences of capturing the glideslope from above. Although some additional training could help pilots to handle this situation, it was agreed that TEMO should not place the aircraft in such conditions when approaching the FAP. In fact, being above the glideslope when headwind is underestimated worsens the situation since altitude deviation grows quicker. This situation was observed in some runs. Moreover, being too high in the glideslope may impede the autopilot to capture it (signal out of the scale) or even worse, capture the secondary signal lobe at 9o.
o All pilots, including the instructor pilot, manifested their concerns about setting 1000 ft at the FCU once cleared for a TEMO descent. This could lead to safety issues when intercepting the glideslope and they all coincide to suggest that this altitude should be, at least, the glideslope minimum intercept altitude.
o It was observed in several runs that pilots had some doubts on whether flaps/slats or gear should be extended, since in current operations they are used to deploy them when they consider it is adequate. In the experiments they were asked to wait until the system told them, unless they considered they had to deploy them sooner in order to ensure the safety of the flight. In some cases pilots were "perturbed" by the fact they were in an experiment and in normal operations they would have acted differently.
o In some cases it was observed that aircraft crew relied too much in automation, since they wanted to deploy flaps/slats or gear earlier but waited for the system to do so, even if noticing too large path deviations. Some pilots even called out something like "let's wait, the system will tell us what to do".
o In this context, it was agreed by all participants that the aircraft crew should know somehow the altitude deviation limits that are allowed. This could prevent the situations where they did not know if it was too early to override TEMO plan (for instance by deploying flaps before) or could help the aircraft crew to anticipate re-plans. In fact, the majority of pilots complained about the fact that it was not clear for them if the system "was aware" of altitude deviations and therefore they were not sure if a re-plan was coming or not. They asked for something similar to the time compliance indications where they could observe the RTA, the ETA and the allowed time tolerance and therefore, anticipate a re-plan when the time deviation approached to the maximum allowed tolerance.
o Regarding this information on vertical deviation tolerances, pilots considered that it could appear in some FMS page (like a "health status" page) or it could be displayed discretely in the ND. They considered that the energy-time box below the PFD was not convenient, since could draw their attention too much. Moreover, they considered that the information regarding time and altitude deviations would be enough for them, with no need to indicate whether the system was re-planning or not, since the plan transitions were acknowledged to be very smooth.
o It was widely claimed that pilots should know whether the system is working properly or not and situations where re-plannings were disabled without warning (runs #x05) were not acceptable at all. As long as the system is working, the majority of pilots expressed they would not even need the information that the system is re-planning ("plan change" indication in the ND), or perhaps it should be displayed in a less visible screen (such as in a FMS page showing some dashes in the ETA field).
o It was agreed that more work is needed to correctly display and differentiate whether the system has failed (is no longer re-planning or tactically acting on the speed) or the system is still alive but the trajectory (RTA) is unable (and perhaps another RTA could be met).
o Some pilots even asked for a functionality allowing them to enable/disable the TEMO algorithm and to "re-arm" the TEMO algorithm after an unable plan for instance.
o As explained before, pilots were not familiar with the speed-on-elevator guidance law and felt uncomfortable when the trajectory started to show deviations on the vertical path. Some of them suggested that the vertical deviation indicator on the PFD (magenta circle on the left of the altitude tape or "yo-yo") could be removed or fixed to a zero deviation, indicating that the "system is taking care of the path". Otherwise, if the vertical deviations are not zero the pilots interpret the path is not correct and that some actions must be done to regain the nominal path.
o In this context, some difficulties from the pilots to understand the strategic guidance concept were observed, since they are used to tactical guidance systems. For example, some pilots asked why the system took too long to re-plan when it was "obvious" that they were deviating from the path. Again, with some extra training and/or indicating the vertical deviation tolerances and/or modifying the "yo-yo" indications this issue could be certainly mitigated.
o Regarding the use of speedbrakes, the majority of pilots thought that they usage should be minimised. Several reasons were given, such for instance to avoid vibrations, uncomfortable flight for passengers, mitigate wear and tear and other maintenance issues, reduce the probability of failure and even to minimise the chances to forget the spoilers deployed (in case they are operated manually).
o For similar reasons, some pilots had concerns on deploying flap/slats at too high speeds (even if within the allowable range of speeds stated in the airplane flight manual).
o From the GRACE control room the engineers observed a very good situational awareness of the crew regarding the use of speedbrakes. Even if they were deployed or retracted automatically, the pilots were always monitoring their status, even calling out their usage and noticing their status at the ECAM and in the ND (thick blue lines).
o Some pilots commented that crew procedures should be revisited since PF "checks" should be done only when the PM makes a call-out and not when the PM performs any action. They consider that the PF must be concentrated flying the aircraft and not monitoring what the PM is doing.
o Some pilots suggested to include a brief checklist of callouts or to implement an auto-feedback to the ATC of compliance with request.

•ASAS-IM feasibility:
o Regarding whether the Select Target and Spacing Instructions in ASAS-IM operations should be combined, pilots had mixed opinions. Some considered that it might be better to reduce the number of options and key strokes, but the majority considered more adequate to separate these actions, as it was done in the experiments.
o The reduction of options when using the ASAS interface is stressed out by some pilots, as in these runs only the Achieve instruction was used but several options where available.
o None of the pilots had previously experience with ASAS and/or IM operations and it was clearly identified that more training on these topics could have saved some problems in the experiments (such as pressing the keys in the incorrect order or acknowledging the spacing to the ATC before TEMO could find a plan, etc.). It should be noted that external intervention from the GRACE control room was required in some runs to reset the FMS or properly load an IM instruction.
o Regarding this last example, it was suggested to improve the HMI in order to indicate the pilot when the plan was being computed preventing him to press "WILCO" in the datalink window before having a plan. In the case of an RTA this is quite obvious since the ETA appears with dashes, but in the case of IM operations this was not clear enough.
o It was also observed that pilots had concerns regarding the responsibilities for spacing. They were not sure if was still the ATC to monitor separation with the preceding aircraft or whether they should take actions if the separation was not kept. Again, some training in ASAS-IM could have prevented these situations.
o All pilots agreed that IM could reduce in the near future the workload on the cockpit and also on the ATC side, especially if these operations come along with CPDL messages.

• New features acceptance (Auto-QNH, Auto-speedbrakeS and PFD/ND CUES):
o A mix of opinions where gathered regarding whether these new features are strictly necessary to ensure the safety and operational feasibility of the TEMO concept. Some pilots consider these features are not that necessary, but are "nice to have", even in manual mode; while others consider that TEMO could only work in automatic flight and providing all these new functionalities are implemented.
o It was pointed out by pilots and engineers that the implementation of some of these new features could seriously delay the introduction of TEMO procedures in real world, due to the big investment needed for certification processes and cost for retrofitting existing aircraft.
o In the big majority of the runs both pilots were aware of the automatic change of the QNH. The manual runs where the ones that reduced the awareness of the pilots in this aspect. .
o Some pilots (Day 3) were very reluctant about the auto-QNH function. They claimed that during the approach the workload is already high and the QNH and altitude check with the ATC increases this workload. Other pilots suggested checking the QNH well in advance or at least not include the altitude check in the radio call with the ATC.
o Nevertheless, the majority of pilots pointed out that the most convenient solution was to reconsider altimetry issues in the TEMO algorithm (re-think for instance how the energy and/or vertical deviations are computed) and maintain current concept of operations regarding transition levels and QNH settings.
o Few pilots said that the auto-speedbrake functionality is not strictly necessary and they could handle speedbrakes manually (providing that the usage is "full/half/zero" and not proportional as done in the experiments). In this context, the algorithm should avoid too many (and small) consecutive segments with the speedbrakes deployed.

•Human Machine Interface Acceptance
o Most of the pilots are comfortable with the TEMO visual indicators displayed and consider they are very useful and necessary. Nevertheless, the fact of showing on the ND each time TEMO is re-planning or calculating a new trajectory is considered to be unnecessary for the majority of subject pilots, since the trajectory changes due to a re-plan might be undetectable from the pilot’s point of view. The general view was that since the re-planning of the trajectory is automatic and running in background, there is no need to display the re-plan notifications (at least on the ND).
o RTA accuracy or spacing information was requested by some pilots to be available either at the PFD or ND display, in order to have a better understanding of the time deviations and RTA adherence.
o The timer that alerts the pilots of the flap/gear configuration change does not start early enough in the opinion of some pilots and they are afraid to not react on time in case they are dealing with other issues (such as talking with ATC). Other pilots, however, thought just the opposite stating that the current timer implementation was good enough and opining that a too long timer time could distract the crew or could not be effective because the crew might deviate their attention from other tasks.
o The energy display used in the manual runs was well appreciated by the pilots and they expressed it was an intuitive way of guidance. Nevertheless, the instructor pilot does not see how this energy indication could actually help the pilots to make the precise corrections when needed if flying in automatic mode (typical operations).
o Some pilots also questioned the visibility of the ATC message eye brown button (size and light intensity). They considered it was not visible enough.
o It was widely suggested to write the ETA in the RTA FMS page with a bigger font size in order to clearly differentiate this time with the early/latest times appearing in the lines above/below.
o As stated before, all pilots missed the information on vertical deviations and their corresponding tolerances during the different phases of flight.
o Regarding the thick lines in the ND, all pilots agree they are absolutely necessary in manual mode, but not essential in automatic mode. The thrust segments, in particular, are not necessary as they consider that when the auto-thrust is working there is no need to show the non-idle thrust segments, since thrust is handled completely in background. Nevertheless, they all agree that having this information in the ND was not a problem and it is a "nice to have" feature.
o On the other hand, most pilots considered that flaps/slats and gear cues on the ND were too big and were concerned about too much cluttering in this display, especially if thinking on airplane types with 4 or even 5 flaps/slats different settings.
o No pilot complained about the different aural timers and few of them considered they were a "must" feature. Moreover, some pilots suggested improving the attention getter.
o All pilots agree that the energy box should not be below the PFD, since it deviates their attention. Most of them found unacceptable to use the amber colour to indicate when the energy/time was out of the limits or even to display the "unable plan" message. This colour is typically associated with serious warnings or emergency situations. As explained before, pilots do want to know the energy/time tolerances (limits before a re-plan is launched) but not in form of this box in the PFD but elsewhere.
o Some pilots suggested a TEMO button in the FCU, perhaps an ENAV (energy navigation) button like the VNAV (vertical navigation).

•Realism of the Simulations:
o All pilots acknowledged the level of realism in the simulations was really high. Being a research experiment, all of them were somehow expecting a less accurate simulation.
o The biggest criticism was for the cockpit configuration, which was a mix of panels, systems and indications from different aircraft types. Some pilots had never flown an Airbus aircraft before and had some difficulties with some indications, while others were confused with the disagreement between the MCP (Boeing-like) and the FMA of the PFD (Airbus-like). Pilots acknowledged that this issue could have an influence on the results, since some errors done (especially in the initial runs) were certainly related with these cockpit layout confusions and lack of training. As a recommendation for future experiments, pilots suggested either to have an exact cockpit reproduction of a specific aircraft type, or to devote some time for a specific training on the new cockpit besides having a manual of the aircraft well in advance.
o All pilots agreed that there was not need to activate the motion in the simulator platform, unless turbulence had to be added and/or system failures to be simulated.
o Some pilots complained about the lack of realism of the operations leading to too low workloads for the aircraft crew. Without the need of going for extreme conditions (thunderstorms or wind-shear) or aircraft anomalies (engine failure...), they considered that some turbulence, instrument meteorological conditions and other traffic were missing in order to have a realistic baseline where to study the benefits of TEMO in terms of workload reduction. In this context, pilots were afraid that some experiment results might be too optimistic because the simulated scenarios were too simple and required a low workload for both of them.
o Some pilots were expecting more demanding scenarios and could not find any differences between some runs (i.e. they did not appreciate that the wind prediction error was increasing in the following run). Some of them suggested that instead of 3kt/6kt/9kt of prediction errors some larger values should have taken.
o One Cessna Citation pilot commented that the average rate of descent observed (around 4000 ft/min) was not acceptable for real operations. According to this pilot, rates of descent for this aircraft type are around 2500ft/min. The main reason explaining this difference is because the nominal descent with the Cessna Citation is not executed with thrust idle. This is an important remark since it will certainly impact on the flight trials planned at the end of the Concorde project.
o Finally, some small remarks and suggestions on how to improve the realism of the simulator were given by some pilots and are summarised below:
▪ the g load indicator should appear only when the load factor exceeds the operational limits
▪ in the ASAS achieve page in the FMS the range field (next to o-clock) should have 1 decimal digit. Moreover, once the target aircraft has landed, this range value should disappear (dashes or blank).
▪ the achieve instruction appears 2 times in the ASAS menu.
▪ Airbus pilots do not like the plan-change and the range-change in the ND. Boeing pilots do not care about it, since range-change is a normal Boeing indication.

*Research Questions

The research questions (RQ) and their corresponding hypothesis (H) were formulated in the experimental plan. In light of the obtained results each research question is discussed below.

RQ-1: Are TEMO CDO in automatic flight considered safe and acceptable by the Subject Pilots, and are the operational requirements met?
H-1: “Subject Pilots accept the arrivals that are flown and find the TEMO procedures safe and acceptable and with no noticeable extra workload. Energy and/or time deviations and stabilization objectives will be, in general, met.”
The hypothesis is not strictly true, since two objectives were not fulfilled related to this question. Nevertheless, the great majority of objectives were successfully achieved and for those that were not achieved only few and very specific runs led to those failures. Therefore it is considered that with some minor improvements this hypothesis could be validated in a future experiment.
As an additional remark, it should be noted that if instead of measuring time performance at the moment the approach mode (APP) is engaged, time deviations were measured at the FAP overfly; some runs would have overflown the FAP with more than +/-5 seconds of error. This is obviously due to the fact that TEMO is disconnected when APP is selected and then, no more re-plannings or active control to nullify time deviations is done. If, eventually, these strict time requirements would be achieved at the FAP, the guidance algorithm should be improved firstly to avoid unable plans when approaching the FAP and secondly to keep nullifying the time/energy error once the APP mode is selected, otherwise the time deviation grows quickly even in very short distances. Perhaps a tactical speed control and a more dynamic flap/slat deployment instructions when approaching the FAP could help to achieve these strict RTAs.

RQ-2: When manually flying with flight director, do Subject Pilots find the TEMO CDO as being safe and acceptable?
H-2: “In these conditions, Subject Pilots are able to successfully fly the TEMO CDO manually satisfying all safety and operational requirements of the procedure. Nevertheless, some performance objectives might not be met due to the flight technical error.”
According to the experiment plan, research Question 2 is considered as an exploratory question since manual flight is not foreseen in TEMO operations. Nevertheless, it was considered worthwhile to assess the 'limits' of the acceptability of TEMO in the hypothetical case manual flight must be performed. Finally, manual runs were always performed at the end of the day and were treated as a "fun-run" for the crew.
For this exploratory research question the hypothesis is not valid. Several pilots considered that TEMO CDO cannot be executed in manual flight. Yet, interestingly the performance objectives are in general met, since the crew devoted a significant effort to manually nullify energy and time errors.

RQ-3: Do Subject Pilots find the TEMO CDO automatic operations acceptable and safe in presence of large forecast errors?
H-3: “Although, in these conditions a significant number of re-plans is expected, Subject Pilots still accept the arrivals that are flown and find the TEMO procedures safe and acceptable.”
The hypothesis is not strictly true, since one objective related to this question was not achieved. Nevertheless, the great majority of objectives were successfully achieved and only few and very specific runs led to its failure. This objective refers to the acceptability of TEMO ASAS-IM procedures. Perhaps more training in TEMO and ASAS-IM operations, along with a more standardised cockpit layout could have help to achieve these objectives. Nevertheless, the interception of the glideslope from above and the lack of awareness regarding the vertical deviation tolerances are identified as potential issues to be further assessed.
It is considered that with some minor improvements this hypothesis could be validated in a future experiment. Moreover some factors external to TEMO may have influenced to the negative results of these two objectives.

* Conclusions

Whilst considering the conclusions to be drawn from the responses to the questionnaires and any issues that have emerged, the over-riding conclusion to be drawn from the experiments themselves is that TEMO is an excellent concept that has a promising future role in ATM facilitation.

There is no doubt that some of the safety issues that have been raised are as a direct (indeed, even indirect) result of problems resulting from the simulator platform itself. The generic nature of the cockpit layout and a performance that was not necessarily well understood by all participating pilots, could have affected the accuracy and reliability of the results.

There was also a strong feeling that the development is at the stage where an ATC involvement in the concept is vital in determining safety and feasibility issues.

The following are the main conclusions that come out from this analysis:

•General improvements proposed for new versions of the TEMO FMS algorithm:
o Avoid the requirement for an auto-QNH implementation and improve the algorithm in such a way that it is robust to current practices regarding the change in altimeter settings.
o Improve the re-planning algorithm to avoid too many consecutive re-plans (improve the look-ahead time, the estimation of the starting point for next plan, etc.).
o Improve TEMO algorithm at low speeds (where wind forecast errors become more important) and especially, when approaching the FAP. These improvements should go in the line to solve the issues related with the interception of the glideslope from above, the difficulty to maintain a RTA at the FAP (leading to some unable plans when approaching the FAP), the multiple and consecutive re-plans when approaching the FAP etc. Possible solutions might be, for instance, to employ a tactical guidance on speed when approaching the FAP, to dynamically schedule flaps/slats transitions, to enforce a minimum time between re-plans, etc.

•Additional measures should be taken to avoid the crew to rely too much in automation:
o The crew shall have a clear idea in which situations they can (or must) override TEMO guidance and act to solve, for instance, a too large vertical deviation. They should also be able to revert to TEMO operations after an eventual override.
o In this context, the crew should know which are the energy/time tolerances that TEMO re-planning algorithm is considering. Another option is not showing vertical deviations in the PFD, indicating the crew that the system is "taking care" of the vertical path too.
o TEMO algorithm works by monitoring time and energy deviations. Energy deviations, however, are difficult to appreciate for the pilots. Thus, before displaying these limits, some conversions to altitude and speed tolerances should be performed.
o If the algorithm is no longer re-planning, the crew should be warned. Nevertheless, in the experiments it has been shown that without this warning they still can handle the situation, which is a good indication from a safety point of view.

•The TEMO CONOPS could be improved:
o ATC clearances shall be revisited, namely the auto-QNH clearance and the descent clearance (in such a way pilots do not select 1000ft at the FCU).
o Crew procedures after an unable plan could be improved and pilots should have the option to re-launch TEMO after a while.
o Maintenance recommendations and operator SOPs, regarding flaps and speedbrake usage should be reviewed considering wear and tear limitations and/or passenger comfort considerations.

The adoption of all or some of these points in future experiments might result in a different set of pilot responses and might address some of the issues; however, that is a subjective assessment and there is no guarantee of such success.

Additional recommendations regarding future experiments:

There were 3 universal responses with regard to any future experiments as follows:
i. The cockpit layout and performance should be realistic and apply to one aircraft design (Airbus?).
ii. There should be a more ‘normal’ workload scenario (more traffic, ATC, IMC, turbulence, etc).
iii. There should be active participation of ATC.

Additional considerations regarding CONCORDE Flight Trials:

o If the flight trail experiment is to produce workable results, then the rate-of-descent of the Cessna cannot be 4000ft/min (query idle descent).
o TEMO planning and guidance algorithm should be tested before flight trials considering non-standard pressure and temperature profiles including some forecast errors in these magnitudes, since this will lead to some energy deviations even at the beginning of the new plan.
o There has to be pre-trial ATC liaison and acceptance and ATC must be thoroughly briefed about the concept.
o There has to be agreement and understanding about when/what descent clearance is given (eg does a TEMO descent need its own clarification vis phraseology – this could take some organising).

SECOND TEMO SIMULATION EXPERIMENT
---------------------------------------------------------------------

TEMO flight simulations were performed at topic manager Braunschweig premises along 3 days: 23rd, 24th and 25th of September 2014. The experiment was focused to investigate the possibility of using the TEMO trajectory prediction on a modern aircraft with unmodified or only slightly modified avionic systems.

This was achieved by running the TEMO concept in an Electronic Flight Bag (EFB) and using the pilot to “close the loop” by entering speed/heading/altitude/vertical-speed instructions into the autopilot Flight Control Unit (FCU).

* Participant remarks:

Below, the remarks that the participants in the simulations made are summarized and grouped into different topics. These include opinions from all pilots involved, the instructor pilot and project engineers. The opinions were gathered from group debriefings and discussions, notes taken during the simulations and remarks that some pilots made directly in the questionnaires.

•TEMO CONOPS and economic feasibility
o All pilots agree that the system could be attractive to airlines; however they do not see it being implemented until ATC promotes a change over the ATM system focused in 4D navigation.
o All agree that, at present, even small savings in fuel (and noise in some airports) are significant enough to justify the certification and adoption costs for such new technology.
o All coincide that this new system is providing good RTA results as long as the time errors at FAP are usually below 5 seconds.
o Pilots agree that, over simulations, they were gradually familiarised with the TEMO concept and felt comfortable flying the new procedures. This acceptability evolution could be also detected over the run questionnaires.
o One of the pilots was convinced that TEMO could reduce crew workload towards ATC communications, as long as the communications would be reduced to a RTA indication at a certain waypoint per part of ATC.
o Besides fuel and noise, the gains in capacity and in reducing ATC workload are also acknowledged. All agree that TEMO can really help in maintain or even increase capacity in dense and complex TMAs.
o Discussions related to the implication and responsibilities of the ATC were brought up in several occasions. Although it is well understood that ATM is out of the scope of CleanSky projects, the active participation of ATC was considered of vital importance in the future developments of the TEMO concept.

•TEMO On-board Integration and HMI Acceptance
o Pilots are comfortable with most of the TEMO visual indicators displayed and consider they are very useful and necessary.
o One pilot found non useful the Gear Down and Flaps information boxes displayed over the EFB, as long as both indicators are already displayed over the Engine Warning and System Displays (EWSD).
o One pilot considered that the function and visual representation of the flap and gear cues were not clear enough, more specifically that they were too small. The same pilot considered that they were actually not necessary to ensure the TEMO operations. The other two pilots though just the opposite answering that these cues were actually necessary.
o All pilots had the feeling that they were in the loop while introducing the EFB indications into the FCU as after several runs all of them tried to anticipate the next EFB indication. Moreover, they all intended to help the system once TEMO was disconnected at the FAF by decreasing the time error produced by the unexpected wind condition.
o The integration of TEMO concept inside an EFB is seen by all the pilots as the most suitable way to integrate TEMO inside the aircraft in the short/mid-term. It would suppose a small inversion for the airlines, in terms of equipment, and the consumption benefits could be noticeable from the first flight.
o All pilots complained about the position of the EFB inside the cockpit. Located in right side of pilot’s position, it was too far from pilot’s eye view prompting him to turn the head every few seconds to check the EFB display. In case of the A320, it was suggested to place the EFB on the retractable table in front of the pilot.
o Related with the time indications on the EFB, one pilot considers the possibility of adding a time trend scale just next to the time indication. By adding this time trend, the pilot considers that the anticipated actions that one can realise would be even more accurate and efficient.

•Operational feasibility and potential safety issues
o The three pilots got a good impression on TEMO’s operational feasibility and considered it to be a possible system from a future aircraft cockpit.
o During the first simulations on each day, pilots were focused on following all the EFB indications strictly, a behaviour that was increasing its workload and hoarding most of their attention due to the numerous speed changes. This situation even caused the pilots to set the QNH or the Missed Approach altitude later than they would normally do it. Then, they started using the EFB indications as suggestions, not to be used exactly but as an indication that pilot is able to consider. This was proved to be the right way of using the system without adding extra workload over the pilot.
o Pilots remarked that they always wanted to reduce the time error zero as soon as possible. However, the EFB indications were not producing big changes in short time, and in some cases this was not understood by pilots.
o In some cases it was observed that pilots relied too much in the EFB alerts, since they wanted to deploy flaps or gear earlier but waited for the system to raise the configuration alert.
o Time error and altitude error indicators were useful to help the pilots to anticipate configuration alerts.
o Regarding the use of speedbrakes, the majority of pilots thought that their usage should be minimised. Several reasons were given, such for instance to avoid vibrations, uncomfortable flight for passengers, mitigate wear and tear and other maintenance issues.
o Some pilots suggested that the EFB indicators update should be triggered using bigger steps of changes in order to attenuate constantly small changes.

•Realism of the Simulations
o All pilots acknowledged the level of realism in the simulations was really high. Being a research experiment, all of them were somehow expecting a less accurate simulation.
o Some pilots complained about the lack of realism of the operations leading to too low workloads for the aircraft crew. Without the need of going for extreme conditions (thunderstorms or wind-shear) or aircraft anomalies (engine failure...), they considered that some turbulence, instrument meteorological conditions and other traffic were missing in order to have a realistic baseline where to study the benefits of TEMO in terms of workload reduction.
o Some pilots were expecting more demanding scenarios and could not find any differences between some runs.

*Validation Objectives:

All validation objectives (VO), stated in the experiment plan, are reviewed below.

VO-1: To assess the feasibility of flying an optimized trajectory by means of speed, heading and altitude/altitude rate commands.
During the first day of simulations, the speed profile used during the TEMO approach was identified by pilot as higher than recommended during a conventional approach. This could be one of the reasons why, the speedbrakes usage was increased and it was identified as not acceptable in one of the trials. This issue was solved by defining a new lower speed profile for next simulation days.
During the third day of simulations, one of the pilots expressed that the usage of the Electronic Flight Bag was not operationally acceptable. Furthermore, it must be noted that the disagreement mark was made at the very first flight simulation performed during the day. In all days we can see clear positive evolution of pilot’s EFB acceptance while going through the different simulations. Then, it can be remarked that this bad grade could be due to lack of confidence with the new equipment that is gained after some practice.

VO-2: To assess whether the provided guidance information are sufficient to guide the aircraft along the calculated trajectory.
During the first day of simulations pilots did complain about the high number of updates that were experiencing the EFB indicators along the procedure. However, during the other days, the Subject pilots had no issues when guiding the aircraft along the trajectory with the information provided. All their subjective answers in each 6 runs have fulfilled the acceptance criterion. The issues that appeared in the first Day were corrected during the 2nd and 3rd day by highlighting more during the training session that the EFB indications were not supposed to be followed automatically by the pilot, they were just that, indications. We consider that this small correction made the same objective to pass during the last two days of simulations.

VO-3: To assess whether the additional task to guide the aircraft along the calculated trajectory is acceptable to the pilot during this phase of flight.
Along the three days, pilots were not totally satisfied with the EFB position and the interval of time that speed changes occurred. Moreover as TEMO automatically was disconnected after arming the APPROACH Mode, the time error increased in the final part of the trajectory. This made the pilots to consider that the system was not working well enough especially in wind scenarios where this time error kept growing once TEMO was disconnected.

VO-4: To assess the additional workload caused by the task to guide the aircraft along the calculated trajectory.
During the first simulations of the first day, bad marks were given related to the workload associated to TEMO. The following simulations were marked as positive, which again indicates a positive evolution of pilot’s EFB acceptance while going through the different simulations.
Regarding the workload rating, it is a reflection of the high amount of speed changes and the wrong conception of the pilot that tried to follow exactly all the EFB indications. Therefore, when asked if he thought that he was successful in accomplishing what he was asked to do he negatively rated this question.
Then, the workload measure of the pilots during the second and third simulation day fulfilled all the acceptance criteria. The pilots encountered no difficulty and no increase in the workload when realizing the tasks demanded.

VO-5: To assess the flight technical error (altitude deviations from the calculated trajectory and time deviations from the RTA).
During all simulation days, both altitude and time deviation along the trajectory have kept between the allowed margins.

* Conclusions:

Among all the results assessed from the responses to the questionnaires and comments from simulations’ participants, the over-riding conclusion to be drawn from the experiments themselves is that TEMO is a concept that has a promising future role in ATM facilitation and that this can be integrated on-board using an EFB using the pilot to close the loop with the aircraft automatic pilot.

The following are the main conclusions that come out from this analysis.

•General improvements proposed for new versions of TEMO in EFB system:
o Improve the system to allow procedures with turns instead of limiting it to straight procedures.
o Improve the system to consider wind and non-standard atmosphere in the trajectory planning process.
o Improve the system in such a way that new re-plans can be performed once the trajectory is being flown (in order fulfil a CTA update and/or cope with too large deviations from nominal plan).
o Improve the indicators update rate, to avoid too many consecutive changes
o Define a better EFB positioning inside the cockpit.

•Plan a set of more realistic simulations:
o Design more realistic scenarios where pilots could test the new system under "normal" workload conditions (turbulence, real atmosphere, surrounding traffic, IMC, ATC communications, etc.)
o There should be active participation of ATC.
o Simulate TEMO EFB failure situations

TEMO ASSESSMENT
---------------------------------

In order to evaluate the TEMO concept with respect to operation/safety and certification aspects by checking that the concept copes with current best practices for CDOs, and that no operational issues arise because of the TEMO function, a concept assessment was performed.

Based on the analysis of TEMO concept and the simulations performed, it can be concluded that TEMO is fulfilling the most relevant aspects of a SGO ITD concept:

• ATC/Pilot Communications: during the simulations, TEMO communication procedures have been proved to be efficient and well accepted by all pilots.
• Pilot Workload: during the simulations, pilots were rapidly familiarised with TEMO concept and felt comfortable flying the new procedures without increasing the workload.
• Predictability to ATC/Pilots: as long as TEMO is based in 4D operations, this results in an increase of predictability enhancing the efficiency of the air traffic.
• Optimal Descent Trajectory: TEMO is always computing the optimal descent trajectory based on the specified restrictions and assuming idle thrust and no speed-brakes used if possible.

However, there are few of them that are still to be tested or improved:

• ATC Workload: The CONCORDE flight simulations were not intended to test the ATC workload, hence there is no data to analyse the ATC workload for managing TEMO operations. In line to cover this aspect, future experiments should be designed considering active participation of ATC.
• Flight Stability: In some simulated flight, the 1000 ft height above the runway was reached without achieving all of the flight stability objectives. In most of these cases the final approach speed objective was not yet fully accomplished. .
• Cost Savings, Environmental Benefits and Noise Reduction: Even if the TEMO concept is designed to cover all these aspects, during the simulation experiments these were not assessed because of the lack of a baseline scenario to compare TEMO operations.

Finally, based on the TEMO detected improvements from simulation, it can be seen that most of them are focused in improving the interaction between flight crew and TEMO, adding new features to the FMS while improving its operation in the real world. However, none of these improvements are affecting at the concept’s fundamental aspects that, from simulation participants point of view, defines TEMO as the best trajectory optimization technology they have ever seen.

TEMO FLIGHT TRIALS
----------------------------------

TEMO flight trials were performed at Groningen Airport Eelde, in the period from 9th till 26th of October 2015. At 9th, 14th and 16th of October, shakedown flights were executed. Measurement flights took place at 19th, 22nd, 23rd and 26th of October 2015. The experiment was focused to demonstrate the TEMO concept in a relevant (real-world) environment. The results of the different flight trials should support the TEMO concept beyond Technological Readiness Level 5 (TRL-5) towards TRL-6.

Hence, the scope of the flight test validation was limited to obtain pilot feedback on the TEMO implementation in various scenarios, with different planning and guidance implementation variants of the concept.

* Participant remarks:

Below, the remarks that the participants in the flight trials made are summarized and grouped into different topics. These include opinions from all pilots involved, the instructor pilot and CONCORDE engineers. The opinions were gathered from group debriefings and discussions, notes taken during the simulations and remarks that some pilots made directly in the questionnaires.

• Operational feasibility and potential safety issues
o Pilots are in general satisfied with the TEMO concept and remark that it is very easy to become familiar with the system.
o In contrast to what happened during the GRACE flight simulations, the aircraft intercepted in the majority of runs the glideslope from above, causing trouble to some pilots who were not used to; in the flight trials TEMO planned trajectories led to interceptions of the glideslope from below in such a way that modelling/guidance uncertainties could lead to too late interceptions (or event to remain always below the glideslope). When that happened, in some cases it was solved by applying manually some pitch up.
o It was remarked that pilots should know what is exactly doing the system in which pilots are relying their guidance. Thus, instead of using different TEMO configurations (combination of tactical and strategic controllers) it should be used a single one which could be understood at each time by flight crew. From both options, strategical controller was better accepted due to the less number of changes.
o In some flights it was faced large vertical speed jumps after TEMO re-plannings, which were difficult to follow immediately by the aircraft. This delta time for achieving the commanded IAS was causing energy errors into the calculated profile leading, sometimes, to a new re-planning. Hence, it was identified the requirement of improving the transition segment the TEMO calculates to move from one energy profile to another after a re-planning.
o The time of flight re-planning was, sometimes, larger than expected (>15sec).
o Regarding TEMO commanded large vertical speed values, the majority of pilots thought that it should be bounded to assure a comfortable flight for passengers all the time. For similar reasons, some pilots show their concerns on using extensively speedbrakes.
o Even if TEMO was accepted by all pilots as a valid concept, it was also raised by one pilot that this TEMO implementation could not be feasible yet in a busy airspace.
o Auto-config and auto speedbrake would be nice to have but is not an absolute requirement.
o Proportional speed brakes would improve performance though and are more important than auto speedbrakes.

• Human Machine Interface Acceptance
o Most of the pilots are comfortable with the TEMO visual indicators displayed and consider they are very useful and necessary.
o Pilots remarked that the timing in TEMO is very critical, so timers are absolutely necessary, unless automatic flap extension is incorporated. During the flight trials the flaps/gear timer aural alert was not working due to a technical problem.
o In general N1 and speed brakes indicators were well followed by the flight crew.
o Altitude and time energy indicators were well appreciated by the pilots and they expressed it was an intuitive way to understand why a plan is working, and when to expect a replan.

* Research Questions:

The research questions (RQ) and their corresponding hypothesis (H) were formulated in the experimental plan. In light of the obtained results each research question is discussed below.

RQ-1: Given the limitations of the platform, can the operational requirements be met?
H-1: “Operational requirements will be met, with the tactical speed guidance implementation showing better time performance, however not all performance objectives will be met due to the platform limitations.”
This hypothesis can be validated if both Time and Stabilization requirements are met. From all of these requirements, just two of them are not fulfilled: time deviation at THR and VCAS deviation at 1000’.
In the first case, to fulfill the time deviation at THR, the time difference between the arrival time and the RTA should be lower than 5 seconds. This requirement is not achieved in most of the runs because, on one hand, once the G/S is intercepted the TEMO SOE controller is disengaged in order to maintain the trajectory path (the G/S) and any modelling or guidance error quickly increases the time error. On the other hand, the aircraft is flying at lower speeds and wind forecast errors are relatively more important. Furthermore, it was also identified that engine dynamics cannot be neglected here, since throttle is used to maintain speed. In this context the planned trajectory assumes instantaneous throttle changes (and thus instantaneous N1 changes).

Regarding the difference between VREF and VCAS at 1000’, 4 TEMO runs (out of 15) are inside the valid formal boundaries (VREF < VCAS < VREF + 5kt). It is also seen that these 4 valid runs are found in the TEMO Energy Strategic/Time Strategic configuration. Main causes of not complying with this requirement are:
• Criteria is too restrictive on lower boundary. In most runs with VCAS < VREF the difference is within 2 kt. This may still be acceptable from an operational point of view.
• Criteria is too restrictive on upper boundary. The tactical controller may command a higher speed than VREF to compensate for a late arrival. This criteria does not take this situation into account.
• The final approach phase was flown manually and the workload for the pilot was rather high (many ‘cues’ to follow). Better results would be expected with an autopilot and/or improved HMI.

RQ-2: Given the limitations of the platform, do Subject Pilots find the TEMO CDO safe and acceptable?
H-2: “In these conditions, Subject Pilots are able to successfully fly the TEMO CDO, satisfying all safety and operational requirements of the procedure. Nevertheless, some performance objectives might not be met due to the platform limitation.”
Considering all pilot feedback given, we can conclude that this research question is fulfilled, given the limitations of the platform, in general subject pilots find the TEMO CDO safe and acceptable, however the flight trials results indicate that the following implementation aspects requires more attention:
• Interception at the ILS glide slope should be better ensured.
• The transition phase requires more tuning. Energy plan seems too dynamic in the current implementation.
• Number of replans should be limited.

RQ-3: Will the TEMO planning and guidance strategy show a fuel benefit over representative current day planning and guidance implementations?
H-3: “While in all implementation the operational requirements will be met, with TEMO however this will be possible at a lower fuel cost.”
This hypothesis can be validated by comparing the fuel consumptions spent while flying with the FMS or TEMO. TEMO runs (in all configuration modes) are reaching lower values of fuel consumption than for the FMS step down runs.

* Conclusions:

As demonstrated in the flight trial report, performing real aircraft flight trials is more complex than performing a high fidelity simulation experiment. Environment conditions are less controlled while there are multiple TEMO implementation and operational aspects that may require tuning during the trial execution. There is no doubt that most of the safety and operational issues encountered, which led to impairment of some requirements as defined in the Experiment Plan, are caused by limitations of the tested TEMO implementation, suboptimal parameter and procedure tuning.

Then, taking into account the results obtained, the following can be concluded:
• TEMO offers a good time performance based on the configured RTAs through all the procedure; but it was observed that on the final segment, in some cases, time deviation increased due to:
o Aircraft speed deviations from the commanded speed while manually flying the glideslope due to a higher pilot workload during the G/S, the engine dynamics or the aircraft performance modelling.
o Wind forecast errors at final.

• Given the limitations of the platform, in general subject pilots find the TEMO CDO safe and acceptable; however the flight trials results indicate that the following implementation aspects requires more attention:
o Interception at the ILS glide slope should be better ensured.
o The transition phase requires more tuning. Energy plan seems too dynamic in the current implementation.
o Number of replans during descent should be limited as long as these could make the whole operation less stable.
o Anti-ice system, and how this affects to aircraft performance, was not taken into account when planning the trajectory.

• TEMO plans the aircraft to be stabilised at 1,000’. However, it has been observed that in some cases the G/S was not captured properly following TEMO guidance.
• TEMO can maintain the high throughout while preserving fuel benefits compared to step-down descents (in line with current day CDA fuel consumption).

Beside these, the following conclusions came out from this analysis.

• Benefits identified for each TEMO configuration variant
It must be taken into account that a very limited number of flights per each TEMO configuration has been performed (Ts/Es: 6, Tt/Es: 4, Ts/Et: 1, Tt/Et: 4). The following outcomes are based on these flights; however, to consolidate them, it will be required to perform further flight campaigns, or a proper batch study using accurate simulation tools, increasing the number of flights in order to have significant statistical values.
o TEMO Ts/Es is the configuration that obtained the better results in speed deviation at 1000’.
o TEMO Ts/Es has been voted by the pilots as the preferred TEMO conceptual variant. It was remarked that strategic controller provided a more stable picture, whether tactical controller cause a lot of speed changes with the related throttle movements.
o TEMO Tt/Et has achieved the lower absolute mean time delta error at THR.
o TEMO Ts/Es has achieved the lower absolute mean time delta error at IAF and at FAP.

• Lessons learnt
o To test various different TEMO conceptual variants can lead to flight crew confusion:
- Which TEMO cues should be followed?
- What is the TEMO system trying to achieve?

This indicates that more crew training is required for these type of trials.

o When testing such a complex system (involving software, procedures design, execution modes, HMI, avionics interface, etc.) a large number of unexpected issues/drawbacks can occur at execution time. To mitigate this risk the following could be done:
- Increase the number of shakedown flights.
- Increase time period in between shakedown flights and measurement flights to enable intermediate data analysis.
- More crew training.

• General improvements of the TEMO FMS algorithm proposed:
o Improve the re-planning algorithm to avoid too many consecutive re-plans (improve the look-ahead time, the estimation of the starting point for next plan, etc.).
o Improve TEMO transition segment after replanning to better fit with the aircraft performance avoiding passenger’s discomfort (large jumps in vertical speeds, IAS changes).
o Make the concept more robust to uncertainties due to weather forecast, aircraft performance modelling, etc.
o Take into account when planning the trajectory the possible usage of the anti-ice system and how this would affect aircraft performance (namely engine idle thrust). A possible solution would be to launch a re-plan each time the anti-ice system is switched on/off.
o Maintain SOE guidance down to the nominal glideslope interception altitude and then implement another strategy (such as an altitude hold mode, or constant flight path angle) to properly capture the glideslope.

In general terms, the TEMO concept has been successfully tested in a real-world environment with positive results and feedback of the participants.
The flight trials have demonstrated that accurate timing can be achieved while preserving fuel benefits in line with current day fuel consumption of CDAs. A promising prospect that indicates that the capacity challenge can be addressed while greening aviation.

CERTIFICATION ASPECTS AND SOLUTIONS FOR TEMO
-------------------------------------------------------------------------------------------

Throughout the CONCORDE project framework, TEMO concept has been successfully tested in different simulator and real experimental flight campaigns in order to demonstrate its operational feasibility and environmental benefits. Moreover, due to the advance planning/guidance features introduced by the concept, its integration and certification on-board a commercial aircraft was identified as a attention point on which some analysis should be carried out.

Inside this framework, the following points can be concluded:

• TEMO integration on-board can be considered through two different approaches:
o Software integration: TEMO algorithm will be integrated in the FMS as an additional software without adding any hardware device. This will require the on-board FMS to fulfil the TEMO hardware specifications.
o Hardware integration: TEMO algorithm will be integrated in an additional flight computer, fulfilling TEMO requirements, interfacing with the on-board FMS.

• In order to calculate the optimum speed trajectory, TEMO will require to receive updated meteorological forecast estimation along the aircraft trajectory. Own aircraft sensor data can be obtained through the FMS, however the actual meteorological forecasts are slowly updated and of low resolution. Hence, it is proposed to integrate a new weather subsystem that could receive through data link channel (SATCOM, VHF, ADS-B) the updated weather forecast data.

• Different equipment from the on-board FMS will need to be modified/added in order to manage TEMO operations:
o FMGC should be modified to understand and transmit TEMO commands to autopilot
o MCDU should be modified to provide control over TEMO parameters and features
o SEC should be modified to accept the new speed brakes configurations commanded by TEMO
o EFIS should be modified to integrate the new cues and tools to aid the pilots while performing TEMO procedure
o FWC should be modified to play new TEMO audible cues for pilots information
o Auto Speedbrakes system should be developed
o AP/FD should be modified to accept new TEMO control conditions

• Due to the required FMS modifications and the inclusion of new systems on-board, TEMO system will be forced to fulfil the following regulation and standards to be certified:
o FAA FAR/EASA CS 23.1309 and 25.1309
o FAA FAR/EASA CS 33.28 and 33.75
o SAE ARP-4754A / EUROCAE ED-79A. Guidelines for the Development of Civil Aircraft and Systems
o SAE ARP-4761 Guidelines and Methods for Conducting the Safety Assessment Process on Civil Airborne Systems and Equipment
o DO-297/ED-124 Integrated Modular Avionics (IMA) Development Guidance and Certification Considerations
o RTCA DO-178C / EUROCAE ED-12C Software Considerations in Airborne Systems and Equipment Certification
o RTCA DO-254 / EUROCAE ED-80 Design Assurance Guidance for Airborne Electronic Hardware
o FAA / EASA TSO-C115c, Flight Management System (FMS) using multi-sensor inputs
o FAA / EASA TSO-C153, Integrated Modular Avionics Hardware elements
o FAA / EASA TSO-C198, Automatic Flight Guidance and Control System (AFGCS) Equipment
o Supplemental Type Certificates

• Because TEMO algorithm is developed using a non-linear programming (NLP) solver for the trajectory prediction and optimisation, TEMO system can be characterized as an adaptive system with nondeterministic feature which conflicts with current civil aviation certification processes, based on the idea that the correct behaviour of a system must be completely specified and verified prior to operation.

• The DO-178C / EUROCAE ED-12C, which defines the Software Considerations in Airborne Systems and Equipment Certification, does not prescribe a specific development process, but instead identifies important activities and design considerations. Some of them could be challenging to be fulfilled by an adaptive system like TEMO:
o Comprehensive requirements: One persistent challenge presented by adaptive systems is the need to define a comprehensive set of requirements for the intended behaviour. The dynamic nature of adaptive systems can make it difficult to specify exactly what they will do at run-time.
o Verifiable requirements: Assuming that a comprehensive set of system requirements can be defined, it may be difficult to verify those requirements by different reasons (availability of verification method, well-defined behaviour, implementation language limitations or structural coverage).
o Documented design: Many certification requirements amount to providing detailed documentation of the system. This may present difficulties for some adaptive systems if they were not initially developed with certification in mind (control of source code, traceability).
o Transparent design: The certification process assumes that a perspicuous, transparent design and software implementation is being presented for evaluation. Since the certification authorities must be convinced that requirements have been met, they must (at least to some reasonable degree) be able to understand the artifacts and evidence that is brought before them. This may be challenging in several ways.

Potential Impact:
DISSEMINATION & COMMUNICATION ACTIVITIES
-------------------------------------------------------------------------------

The aim of CONCORDE was to evaluate the viability of the Time and Energy Managed Operations (TEMO) concept, developed inside the Systems for Green Operation (SGO) ITD framework. More specifically, the aim of CONCORDE project was to help define, prepare, perform and analyse two flight simulator experiments, which were a prerequisite to perform a flight test of the TEMO concept.

Even if there were no specific work package with dissemination activities defined, CONCORDE consortium got engaged to disseminate the results and evolution of the project within the aviation community, in line with the requests of CSJU.

* Key message:

The dissemination activities developed in the framework of CONCORDE aimed to announce the following key messages:
• Project purpose and development: it was important to transmit which was the aim of CONCORDE project and its value on testing the TEMO concept.
• TEMO TLR-5 achievement: one of the main project objectives was to perform the required simulations to prove that TEMO concept has successfully reached TRL-5.
• TEMO trials performance: after reaching TRL-5, there were executed successfully a set of flight trials demonstrating the TEMO concept in the real world.

* Target audiences:

Publications have been addressed to specific audiences with the help of the specific channels selected. There were three main target audiences of the project. Firstly, the general public, secondly, the technical community that involves research centres and universities and finally industrial stakeholders, endorsers or clusters that may be interested in the activities of the Project.

* Communication activities

• Communication through Partners Websites:
During the Project, consortium members published news about the Project updates and milestones, to keep interaction with the target audiences and generate awareness of the activities and objectives of the Project.

Concrete publications by the consortium members included:
- Press release created by Pildo Labs to present Project
- News in Clean Sky website
- News in UPC website at the beginning and end of the Project.

• Scientific papers:

Two scientific papers, based on the work performed in CONCORDE project, have been published, and a third one is on the way. The following sections contain the abstract of the papers:

- Evaluation of in-flight trajectory optimisation with time constraints in a moving base flight simulator.

This paper describes an aircraft trajectory optimisation algorithm that is part of the time and energy managed operations (TEMO) concept, a new trajectory planning and guidance strategy for advanced four dimensional (4D) continuous descent operations. In TEMO, a trajectory is planned in such a way that the best vertical trajectory profile is computed (according to a given optimisation objective), while satisfying at the same time
air traffic control required times or arrival (RTA) at one or more waypoints. The guidance part of TEMO is mainly based on a speedon-elevator law, where the calibrated speed plan is followed. TEMO was evaluated in a highly realistic environment with qualified pilots. Results show that the TEMO algorithm largely meets time accuracy requirements at the initial approach fix, while further work is needed to achieve the very demanding accuracies imposed at the runway threshold.
This paper was presented at the 34th Digital Avionics Systems Conference 2015

- Optimization of the vertical trajectory through Time and Energy management: A Human in the-Loop Study.

A new integrated planning and guidance concept has been developed that attempts to optimize the vertical trajectory to achieve a continuous engine-idle descent while satisfying time constraints. The new concept, named Time and Energy Managed Operations (TEMO), reduces noise, gaseous emissions and fuel consumption, while adhering to time constraints yielding control points for spacing and sequencing. The concept allows small deviations between planned and actual trajectory within pre-set boundaries and calculates a new trajectory when these deviations are exceeded. The concept was evaluated in a Full Motion Flight Simulator. Six pilots were invited to provide feedback of the TEMO concept with respect to operation, safety and certification requirements and to demonstrate that the TEMO Technology Readiness Level (TRL) achieved level 5, in that the technology is demonstrated in a relevant (simulated) environment. Conclusions drawn from the responses to the questionnaires and any issues that have emerged, indicate that TEMO is an excellent concept that has a promising future role in ATM facilitation. Recommendations for future research include the use of more involved scenarios and active participation of ATC.
This paper was presented at the AIAA Science and Technology Forum and Exposition 2016

- Optimal continuous descent operations supported by an electronic flight bag: a human-in-the-loop study

This paper describes a set of flight simulation experiments carried out in a simulator. A new trajectory planning and guidance strategy for advanced 4D continuous descent operations (CDO) was evaluated after three full days of experiments with qualified pilots .
The experiment focused to investigate the possibility of using time and energy managed operations (TEMO) trajectory planning and guidance concepts on a modern aircraft with unmodified or only slightly modified avionic systems. This was achieved by running the advanced and novel trajectory planning algorithms in an Electronic Flight Bag (EFB) and using the pilot to “close the loop” by entering speed/heading/altitude/vertical-speed instructions into the autopilot Flight Control Unit (FCU).
In TEMO, the trajectory is planned using a mathematical optimization algorithm that is capable to compute the optimal vertical profile of the trajectory while satisfying Air Traffic Control (ATC) time constraints at one or more waypoints. The aim of TEMO is to allow CDOs in dense terminal areas without compromising the capacity, expecting in this way, to facilitate flow management and arrival spacing increasing the arrival throughput while reducing the environmental fingerprint.
The outcome of the experiments include subjective (questionnaires answered by pilots) and objective (trajectory logs) data. Data analysis showed a very good acceptance (both in terms of safety and operability of the procedure) from the participating crews, only with minor suggestions to be improved in future versions of the TEMO module. Moreover, the experiments showed good time accuracies at the 4D metering fixes that were within the very demanding tolerances proposed in SESAR or NextGen programs.
This paper is going to be submitted to the 35th Digital Avionics Systems Conference (DASC)

* Press release:

A press release was drafted at the beginning of the project to present the project objectives and activities. The press release was distributed to the consortium members and main stakeholders.

List of Websites:
Concorde Consortium Main Contacts:
• Brent Day (Pildo Labs) – land: 01963 364580 – email: brent.day@pildo.com
• Josep Montolio (Pildo Labs) – land: +34 931828845 – email: josep.montolio@pildo.com
• Xavier Prats (UPC) – land: +34 934134125 - email: xavier.prats@upc.edu

Related information

Reported by

PILDO CONSULTING SL
Spain
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top