Periodic Reporting for period 2 - TransAID (Transition Areas for Infrastructure-Assisted Driving)
Période du rapport: 2019-03-01 au 2021-02-28
As the introduction of automated vehicles becomes feasible, even in urban areas, it is necessary to investigate their impacts on traffic safety and efficiency. This is particularly true during the early stages of market introduction, where automated vehicles of all SAE levels, connected vehicles (able to communicate via V2X), and conventional vehicles share the same roads with varying penetration rates.
There will be areas and situations on the roads where high automation can be granted, and others where it is not allowed or not possible due to missing sensor inputs, high complexity situations, etc. At the border of such areas many automated vehicles will change their level of automation. We refer to these areas as “Transition Areas”. It is very likely that Transition Areas will have a negative impact on traffic safety and efficiency, since transitions of control may fail or lead to non-optimal behavior.
TransAID develops and demonstrates traffic management procedures and protocols to enable smooth coexistence of automated, connected, and conventional vehicles, especially at Transition Areas. A hierarchical approach is followed where control actions are implemented at different layers including centralized traffic management, infrastructure, and vehicles.
To achieve this, modelling of the existing and upcoming automation functionalities and driver behavior has been done as a first step, including modelling of transitions of control as well as automated lane changes and speed/distance keeping. This is a mandatory step to get insights not only into current situations, but also to predict the future impact. Therefore, the models have been used to perform simulations in different shares and in a set of 10 use cases. The simulations proved that there is a negative impact on safety, efficiency and CO2 emissions caused by transitions of control and minimum risk maneuvers. By following the hierarchical approach, infrastructure-assisted management solutions have been investigated and simulated advising connected, automated, and conventional vehicles at Transition Areas, taking into account traffic safety and efficiency metrics. These solutions aimed at the prevention of transitions of control and minimum risk maneuvers by providing speed, distance, lane or path information. Also in focus was the distribution of transitions of control along the road. In case a transition of control is unavoidable, the developed solution supported connected automated vehicles to find a safe spot for performing a minimum risk maneuver.
As cooperation between infrastructure and connected/automated vehicles is a key requirement, V2X communication protocols using ITS-G5 have been developed, including collective perception (CPM) and Maneuver Coordination (MCM). Measures to detect and inform conventional vehicles are also addressed.
The most promising solutions have been implemented as real world prototypes and demonstrated under real world conditions, on test tracks and on a highway in the Netherlands. The implementation included road side components (sensors, including object detection and tracking, variable message signs) as well as vehicle components (automation function, HMI) and the communication. Feasibility assessments have been performed using these prototypes, and their results have been used to further enhance the traffic management solutions.
Finally, guidelines for advanced infrastructure-assisted driving have been formulated. These guidelines also include a roadmap defining activities and needed upgrades of road infrastructure in the upcoming fifteen years in order to guarantee a smooth coexistence of conventional, connected, and automated vehicles.
There will be areas and situations on the roads where high automation can be granted, and others where it is not allowed or not possible due to missing sensor inputs, high complexity situations, etc. At the border of such areas many automated vehicles will change their level of automation. We refer to these areas as “Transition Areas”. It is very likely that Transition Areas will have a negative impact on traffic safety and efficiency, since transitions of control may fail or lead to non-optimal behavior.
TransAID develops and demonstrates traffic management procedures and protocols to enable smooth coexistence of automated, connected, and conventional vehicles, especially at Transition Areas. A hierarchical approach is followed where control actions are implemented at different layers including centralized traffic management, infrastructure, and vehicles.
To achieve this, modelling of the existing and upcoming automation functionalities and driver behavior has been done as a first step, including modelling of transitions of control as well as automated lane changes and speed/distance keeping. This is a mandatory step to get insights not only into current situations, but also to predict the future impact. Therefore, the models have been used to perform simulations in different shares and in a set of 10 use cases. The simulations proved that there is a negative impact on safety, efficiency and CO2 emissions caused by transitions of control and minimum risk maneuvers. By following the hierarchical approach, infrastructure-assisted management solutions have been investigated and simulated advising connected, automated, and conventional vehicles at Transition Areas, taking into account traffic safety and efficiency metrics. These solutions aimed at the prevention of transitions of control and minimum risk maneuvers by providing speed, distance, lane or path information. Also in focus was the distribution of transitions of control along the road. In case a transition of control is unavoidable, the developed solution supported connected automated vehicles to find a safe spot for performing a minimum risk maneuver.
As cooperation between infrastructure and connected/automated vehicles is a key requirement, V2X communication protocols using ITS-G5 have been developed, including collective perception (CPM) and Maneuver Coordination (MCM). Measures to detect and inform conventional vehicles are also addressed.
The most promising solutions have been implemented as real world prototypes and demonstrated under real world conditions, on test tracks and on a highway in the Netherlands. The implementation included road side components (sensors, including object detection and tracking, variable message signs) as well as vehicle components (automation function, HMI) and the communication. Feasibility assessments have been performed using these prototypes, and their results have been used to further enhance the traffic management solutions.
Finally, guidelines for advanced infrastructure-assisted driving have been formulated. These guidelines also include a roadmap defining activities and needed upgrades of road infrastructure in the upcoming fifteen years in order to guarantee a smooth coexistence of conventional, connected, and automated vehicles.
TransAID started by investigating future automation functionalities and system behavior, focusing on automation limits. As result, TransAID was investigating infrastructure services which allow vehicles to solve situations which they cannot handle on their own, e.g. in case a vehicle is not able to follow a given route without braking rules, infrastructure may suggest a way to overcome the obstacle. In case a vehicle needs to perform a minimum risk maneuver when the handover of control to the driver fails, infrastructure may suggest a good position for coming to a safe stop. In total, five different categories have been identified.
TransAID investigated use cases and scenarios inside the five categories in two project iterations.
As a first step, the general behavior of automated vehicles, including transitions of control, has been modelled. Simulations using SUMO have been performed showing big impacts of automated vehicles as described above.
As a next step, the identified service use cases have been implemented in simulation, including the traffic management algorithm and the communication protocols and message sets. The latter has been done in line with current V2X standardization activities and does also include measures to detect and advice conventional vehicles, e.g. by using variable message signs.
In order to allow precise simulations, the already existing iTETRIS simulation platform has been chosen as basis, which allows coupling of SUMO simulations with the ns-3 communication simulator. This platform has been used to simulate all use cases.
Afterwards, the proposed traffic management measures have been implemented in real world prototypes on test tracks, including behavior of automated vehicles, communication and infrastructure components.
The complete cycle has been repeated during the second project iteration. Therefore, the use cases have been refined and combined. Again, simulations have been performed using SUMO and iTETRIS, before implementing the adapted measures in real world prototypes. Here, three real world feasibility assessments have been performed on test tracks and a public highway.
All results have been integrated in a guideline for road authorities and stakeholders for a smooth introduction of automated vehicles. This guideline is also including a roadmap revealing required steps. Besides numerous publications, the TransAID simulation software has been released Open Source.
TransAID investigated use cases and scenarios inside the five categories in two project iterations.
As a first step, the general behavior of automated vehicles, including transitions of control, has been modelled. Simulations using SUMO have been performed showing big impacts of automated vehicles as described above.
As a next step, the identified service use cases have been implemented in simulation, including the traffic management algorithm and the communication protocols and message sets. The latter has been done in line with current V2X standardization activities and does also include measures to detect and advice conventional vehicles, e.g. by using variable message signs.
In order to allow precise simulations, the already existing iTETRIS simulation platform has been chosen as basis, which allows coupling of SUMO simulations with the ns-3 communication simulator. This platform has been used to simulate all use cases.
Afterwards, the proposed traffic management measures have been implemented in real world prototypes on test tracks, including behavior of automated vehicles, communication and infrastructure components.
The complete cycle has been repeated during the second project iteration. Therefore, the use cases have been refined and combined. Again, simulations have been performed using SUMO and iTETRIS, before implementing the adapted measures in real world prototypes. Here, three real world feasibility assessments have been performed on test tracks and a public highway.
All results have been integrated in a guideline for road authorities and stakeholders for a smooth introduction of automated vehicles. This guideline is also including a roadmap revealing required steps. Besides numerous publications, the TransAID simulation software has been released Open Source.
As mentioned, future automated driving capabilities and weaknesses are not yet well known. Nevertheless, TransAID found traffic management measures which will be able to help automated vehicles in the future in case a locally bound problem occurs leading to a transition of control or a situation which cannot be handled by the automated vehicles alone.
Although the exact kind of problem an automated vehicle may have in the future can only be guessed, it is very important to address solutions for general and very likely problems already very early, which is pioneered by TransAID. Changing the perspective from the inside, just being passenger of an automated vehicle enjoying the increase of comfort, to the outside, where the behavior of automated vehicles impacts the traffic safety and efficiency of all vehicles, is unavoidable. This is especially true when looking at Transition Areas where negative impacts are very likely. By following this approach, TransAID offers solutions for future mobility issues caused by automated vehicles. This includes the early standardization of solutions (e.g. V2X message sets), the early response of stakeholders, etc. – mandatory steps for future smooth coexistence of conventional and automated vehicles.
Although the exact kind of problem an automated vehicle may have in the future can only be guessed, it is very important to address solutions for general and very likely problems already very early, which is pioneered by TransAID. Changing the perspective from the inside, just being passenger of an automated vehicle enjoying the increase of comfort, to the outside, where the behavior of automated vehicles impacts the traffic safety and efficiency of all vehicles, is unavoidable. This is especially true when looking at Transition Areas where negative impacts are very likely. By following this approach, TransAID offers solutions for future mobility issues caused by automated vehicles. This includes the early standardization of solutions (e.g. V2X message sets), the early response of stakeholders, etc. – mandatory steps for future smooth coexistence of conventional and automated vehicles.