Periodic Reporting for period 1 - Multi-Moby (Safe, Secure, High Performing Multi-Passanger and Multi-Commercial Uses Affordable EVs)
Période du rapport: 2020-12-01 au 2022-05-31
Multi-Moby unites and improves the work from a cluster of past European H2020 projects, and aims to develop technology for safe, efficient and affordable urban electric vehicles with numerous objectives defined to maximise the output of this project:
Develop a multi-purpose electric vehicle (EV) fleet, which incorporates the following objectives:
- Demonstrate high passive safety for vehicle occupants and vulnerable road users (VRUs), enhanced by active safety features
- Advance towards autonomous-capable vehicles, applying a step-by-step approach from advanced driving assistance systems (ADAS) to conditional and full autonomy
- Develop advanced energy storage and efficient charging at low and mid-low voltage
- Develop cost-effective powertrains and zone partitioned Electrical Electronic (EE) architecture
Develop a multi-purpose electric vehicle (EV) fleet, which incorporates the following objectives:
- Demonstrate high passive safety for vehicle occupants and vulnerable road users (VRUs), enhanced by active safety features
- Advance towards autonomous-capable vehicles, applying a step-by-step approach from advanced driving assistance systems (ADAS) to conditional and full autonomy
- Develop advanced energy storage and efficient charging at low and mid-low voltage
- Develop cost-effective powertrains and zone partitioned Electrical Electronic (EE) architecture
Multi-Moby has progressed well, with numerous activities and alignments towards the objectives with promising initial results.
Prototypes of the Multi-Moby EVs have been constructed. To meet the ambitious Euro NCAP 4-stars crash standards, several activities related to the vehicle’s passive safety were performed:
Structure optimisation to comply with the requirements of all the frontal and lateral impact regulations and also with more critical Euro NCAP protocols: The first iterative optimisation has been finished, and three regulation crash tests with the prototype vehicles have been performed with positive results. The crash tests measurements have been used to improve the simulation model and a new optimisation loop is ongoing. The simulation and experimental results confirm a good structure performance for all the crash test scenarios considered.
Development of customised restraint systems to fulfil the specific requirements of small-sized vehicles.
For the VRUs’ protection, the design of the frontal vehicle part is being optimised to reduce the injuries to VRUs in the event of an accident.
Optimising the vehicle fatigue behaviour, which has been achieved virtually thanks to the design iterations through the finite element method and experimentally.
With regards to active safety, two active safety features were developed. The first feature is referred to as pre-emptive traction control, in which the information of the expected tyre-road friction coefficient profile ahead, coming from V2X connectivity based on the estimation outputs of preceding vehicles, is used by the non-linear model predictive control (NMPC) wheel slip controller of the ego vehicle to pre-emptively reduce the torque demand to prevent longitudinal slip ratio oscillations, and to compensate for the powertrain actuator delays.
The second feature is referred to as pre-emptive braking control. The V2X information on the expected road curvature ahead is sent to an NMPC braking controller, which pre-emptively slows down the vehicle by controlling its torque demand to ensure desirable levels of agility and sideslip angle in limit handling conditions, without the need for costly chassis actuators. Both sets of active safety features have been tested in simulation and experimentally on the vehicles with promising results, which has been detailed in scientific papers.
With regards to autonomous driving capabilities, the Mobileye 8 ADAS has been installed and configured on one of the prototype EVs. The Mobileye system unites collision avoidance technology with cloud connectivity, fleet platform capability, large detection range, improvements in accuracy, wide-angle camera lens and a g-sensor to detect harsh braking, accelerations and cornering.
Multiple gimbals have been installed on the vehicle for autonomous driving. The high accuracy gimbals enhance sensing and detection of objects and offers precision motion enabling and unique optical zoom and staring capabilities, and can partially replace existing LiDAR sensors. The gimbal has been integrated and optically tested. In addition, vibration suppression performance has been tested and 3D measurement by two gimbals was demonstrated.
In terms of the energy storage system, a novel hybrid supercapacitor-battery in a single cell has been developed, which allows a high peak power and very fast charging. The cells have been appropriately sized for the Multi-Moby vehicle based on simulations for the required power and energy. The placement of the novel hybrid cell packs was developed in a CAD environment and cells are placed in the battery bay of the vehicle. The charging concept has been developed to charge the proposed two voltages in the energy storage systems. A wall box charger was developed, with a high efficiency aim.
With regards to the motorised axle system, the cost-effective powertrain solutions were developed and adapted to the EVs. Both the mid-low voltage powertrain and low voltage powertrain with a belt motor system were provided and evaluated for mounting into the vehicle. Supply of the two inverters for front and rear e-motors has been carried out. The EE architecture has been studied and defined according to the architecture diagram.
Prototypes of the Multi-Moby EVs have been constructed. To meet the ambitious Euro NCAP 4-stars crash standards, several activities related to the vehicle’s passive safety were performed:
Structure optimisation to comply with the requirements of all the frontal and lateral impact regulations and also with more critical Euro NCAP protocols: The first iterative optimisation has been finished, and three regulation crash tests with the prototype vehicles have been performed with positive results. The crash tests measurements have been used to improve the simulation model and a new optimisation loop is ongoing. The simulation and experimental results confirm a good structure performance for all the crash test scenarios considered.
Development of customised restraint systems to fulfil the specific requirements of small-sized vehicles.
For the VRUs’ protection, the design of the frontal vehicle part is being optimised to reduce the injuries to VRUs in the event of an accident.
Optimising the vehicle fatigue behaviour, which has been achieved virtually thanks to the design iterations through the finite element method and experimentally.
With regards to active safety, two active safety features were developed. The first feature is referred to as pre-emptive traction control, in which the information of the expected tyre-road friction coefficient profile ahead, coming from V2X connectivity based on the estimation outputs of preceding vehicles, is used by the non-linear model predictive control (NMPC) wheel slip controller of the ego vehicle to pre-emptively reduce the torque demand to prevent longitudinal slip ratio oscillations, and to compensate for the powertrain actuator delays.
The second feature is referred to as pre-emptive braking control. The V2X information on the expected road curvature ahead is sent to an NMPC braking controller, which pre-emptively slows down the vehicle by controlling its torque demand to ensure desirable levels of agility and sideslip angle in limit handling conditions, without the need for costly chassis actuators. Both sets of active safety features have been tested in simulation and experimentally on the vehicles with promising results, which has been detailed in scientific papers.
With regards to autonomous driving capabilities, the Mobileye 8 ADAS has been installed and configured on one of the prototype EVs. The Mobileye system unites collision avoidance technology with cloud connectivity, fleet platform capability, large detection range, improvements in accuracy, wide-angle camera lens and a g-sensor to detect harsh braking, accelerations and cornering.
Multiple gimbals have been installed on the vehicle for autonomous driving. The high accuracy gimbals enhance sensing and detection of objects and offers precision motion enabling and unique optical zoom and staring capabilities, and can partially replace existing LiDAR sensors. The gimbal has been integrated and optically tested. In addition, vibration suppression performance has been tested and 3D measurement by two gimbals was demonstrated.
In terms of the energy storage system, a novel hybrid supercapacitor-battery in a single cell has been developed, which allows a high peak power and very fast charging. The cells have been appropriately sized for the Multi-Moby vehicle based on simulations for the required power and energy. The placement of the novel hybrid cell packs was developed in a CAD environment and cells are placed in the battery bay of the vehicle. The charging concept has been developed to charge the proposed two voltages in the energy storage systems. A wall box charger was developed, with a high efficiency aim.
With regards to the motorised axle system, the cost-effective powertrain solutions were developed and adapted to the EVs. Both the mid-low voltage powertrain and low voltage powertrain with a belt motor system were provided and evaluated for mounting into the vehicle. Supply of the two inverters for front and rear e-motors has been carried out. The EE architecture has been studied and defined according to the architecture diagram.
For active safety, the next steps will be an evaluation of the influence of the autonomous capabilities of the vehicles, in the safety performance of the occupants and of VRUs. In addition, the gimbals will be combined with the latest generation of ADAS devices with AI-based on-board decision-making processes and will be compared to the Mobileye system.
The zone partitioned EE architecture is intended as a Distributed Learning Framework (DLF) that allows individual autonomous vehicles to upload in real-time or on a time increment basis, their operation parameters (e.g. time of travel, captured images of obstruction, faults and operations failure modes) to a central knowledge base that can be used for continuous training of machine learning models.
In terms of the energy storage system, electrothermal stability will be addressed to assure safety, robustness and an acceptable range with good performance in all climate conditions. The next generation of hybrid supercapacitor-battery cell technology will be integrated into the vehicle. Multi-Moby will also address the optimisation of charging at low voltages, studying and promoting products capable of operating at 200V-250 V (rather than 500 V). This could also allow smaller electrical connectors more suitable for urban EVs.
The use of SiC MOSFET components to improve power conversion performance or implement system innovation is nowadays a popular scenario for many system designers. In fast DC EV charging, 1200 V SiC MOSFET technology enables shortened charging times. Compared to a silicon-based solution, output power can be doubled even with the same footprint thanks to reduced part count and 50% loss reduction, thereby also cutting charging time in half.
The final plan in Multi-Moby is to introduce robot food and medical delivery vehicles to the market one year after the completion of Multi-Moby. The gimbals for autonomous driving is expected to give a technological and price advantage that will facilitate the uptake of affordable self-driving system kits, expected to be available by 2025 at 5000€ or below. This will enable Multi-Moby to address the growing need of a reliable, affordable sensing and high-speed computational on-board platforms, to improve flexibility and optimisation of manufacturing processes and to present multi-purpose vehicles at an affordable cost.
The zone partitioned EE architecture is intended as a Distributed Learning Framework (DLF) that allows individual autonomous vehicles to upload in real-time or on a time increment basis, their operation parameters (e.g. time of travel, captured images of obstruction, faults and operations failure modes) to a central knowledge base that can be used for continuous training of machine learning models.
In terms of the energy storage system, electrothermal stability will be addressed to assure safety, robustness and an acceptable range with good performance in all climate conditions. The next generation of hybrid supercapacitor-battery cell technology will be integrated into the vehicle. Multi-Moby will also address the optimisation of charging at low voltages, studying and promoting products capable of operating at 200V-250 V (rather than 500 V). This could also allow smaller electrical connectors more suitable for urban EVs.
The use of SiC MOSFET components to improve power conversion performance or implement system innovation is nowadays a popular scenario for many system designers. In fast DC EV charging, 1200 V SiC MOSFET technology enables shortened charging times. Compared to a silicon-based solution, output power can be doubled even with the same footprint thanks to reduced part count and 50% loss reduction, thereby also cutting charging time in half.
The final plan in Multi-Moby is to introduce robot food and medical delivery vehicles to the market one year after the completion of Multi-Moby. The gimbals for autonomous driving is expected to give a technological and price advantage that will facilitate the uptake of affordable self-driving system kits, expected to be available by 2025 at 5000€ or below. This will enable Multi-Moby to address the growing need of a reliable, affordable sensing and high-speed computational on-board platforms, to improve flexibility and optimisation of manufacturing processes and to present multi-purpose vehicles at an affordable cost.