Periodic Reporting for period 3 - PRYSTINE (Programmable Systems for Intelligence in Automobiles)
Okres sprawozdawczy: 2020-05-01 do 2021-10-31
The automation of vehicles – ultimately aiming at fully autonomous driving – has been identified as one major enabler to master the Grand Societal Challenges “Individual Mobility” and “Energy Efficiency”. Highly automated driving functions (ADF) are one major step to be taken.
One of the major challenges to successfully realizing highly automated driving is the step from SAE Level-2 (Partial automation) to SAE Levels-3 (Conditional automation) and above. At level-3, the driver remains available as a fallback option in the event of a failure in the automation chain, or if the ADF reaches its operational boundaries. At higher levels, the driver cannot be relied upon to intervene in a timely and appropriate manner, and consequently, the automation must be capable of handling safety-critical situations on its own.
For this, fail-operational behavior is essential in the sense, plan, and act stages of the automation chain. PRYSTINE's main target is to realize Fail-operational Urban Surround perceptION (FUSION), which is based on robust Radar and LiDAR sensor fusion, and control functions in order to enable safe automated driving in urban and rural environments.
The PRYSTINE project's strategy has consisted in achieving its main target through reaching four key technical objectives. While these four technical objectives address different levels (components, control systems, architectures and function) of the automation chain, PRYSTINE has also reached 2 non-technical objectives addressing market/social/technological impacts and “European Values”.
The objectives were:
Objective 1: Enhanced reliability and performance, cost and power of FUSION components
Objective 2: Dependable embedded control by co-integration of signal processing and AI approaches for FUSION
Objective 3: Optimized E/E architecture enabling FUSION-based automated vehicles
Objective 4: Fail-operational systems for urban and rural environments based on FUSION
Objective 5: Competitive advantage for European industry
Objective 6: Increased user acceptance of automated driving functions
In conclusion, the project has reached all its initial objectives and has been able to prove its achievement in a qualitative and quantitative manner.
One of the major challenges to successfully realizing highly automated driving is the step from SAE Level-2 (Partial automation) to SAE Levels-3 (Conditional automation) and above. At level-3, the driver remains available as a fallback option in the event of a failure in the automation chain, or if the ADF reaches its operational boundaries. At higher levels, the driver cannot be relied upon to intervene in a timely and appropriate manner, and consequently, the automation must be capable of handling safety-critical situations on its own.
For this, fail-operational behavior is essential in the sense, plan, and act stages of the automation chain. PRYSTINE's main target is to realize Fail-operational Urban Surround perceptION (FUSION), which is based on robust Radar and LiDAR sensor fusion, and control functions in order to enable safe automated driving in urban and rural environments.
The PRYSTINE project's strategy has consisted in achieving its main target through reaching four key technical objectives. While these four technical objectives address different levels (components, control systems, architectures and function) of the automation chain, PRYSTINE has also reached 2 non-technical objectives addressing market/social/technological impacts and “European Values”.
The objectives were:
Objective 1: Enhanced reliability and performance, cost and power of FUSION components
Objective 2: Dependable embedded control by co-integration of signal processing and AI approaches for FUSION
Objective 3: Optimized E/E architecture enabling FUSION-based automated vehicles
Objective 4: Fail-operational systems for urban and rural environments based on FUSION
Objective 5: Competitive advantage for European industry
Objective 6: Increased user acceptance of automated driving functions
In conclusion, the project has reached all its initial objectives and has been able to prove its achievement in a qualitative and quantitative manner.
PRYSTINE has validated the achievement of all aforementioned objectives through showcasing a set of Key Performance Indicators (KPIs) via a number of demonstrators implemented and shown during all project reviews.
Objective 1:
• 25% less data communication required compared to state-of-the-art
• 30% less false-positive detections compared to separate sensing approach
• Fail operational sensor compound vs. fail silent individual sensing approaches
• Power reduction of 25% through semiconductor material improvements and functional convergence in sensor modules
• Up to 30% cost reduction and 10% margin improvement for perception sub-systems
Objective 2:
• Fail operational system approach for ADF (SAE Level 3+) vs. fail silent advanced driver assistance system (ADAS) approaches (SAE Levels up to 2)
• Proposed Certification Approach for AI based sensor fusion, diagnostic, and control
Objective 3:
• Fail operational system architecture demonstrator for ADF (SAE Level 3+) vs. fail silent ADAS approaches (SAE Levels to 2)
• LiDAR / RADAR sensor compound demonstrator
Objective 4:
• Fail-operational SAE Level 3+ ADFs in urban environments
Objective 5:
• Increased market share and revenue of European companies through PRYSTINE’s ground¬breaking technological advancements (O1-O4)
Objective 6:
• Increased user acceptance of automated driving functions through PRYSTINE’s groundbreaking technological advancements (O1-O4)
Moreover, PRYSTINE has followed a very concrete dissemination, communication and exploitation strategy, that has resulted in the following:
a) 30 project posters
b) 4 leaflets/flyers
c) 6 e-newsletters
d) 20 open-aire publications
e) 105 journal and conference publications
f) participation in total in 81 events
g) ~100 projects exploitable results were extracted from the questionnaires and can be categorized in three major categories according to its type: Knowledge or know-how, software and hardware components, and services or methodologies
Objective 1:
• 25% less data communication required compared to state-of-the-art
• 30% less false-positive detections compared to separate sensing approach
• Fail operational sensor compound vs. fail silent individual sensing approaches
• Power reduction of 25% through semiconductor material improvements and functional convergence in sensor modules
• Up to 30% cost reduction and 10% margin improvement for perception sub-systems
Objective 2:
• Fail operational system approach for ADF (SAE Level 3+) vs. fail silent advanced driver assistance system (ADAS) approaches (SAE Levels up to 2)
• Proposed Certification Approach for AI based sensor fusion, diagnostic, and control
Objective 3:
• Fail operational system architecture demonstrator for ADF (SAE Level 3+) vs. fail silent ADAS approaches (SAE Levels to 2)
• LiDAR / RADAR sensor compound demonstrator
Objective 4:
• Fail-operational SAE Level 3+ ADFs in urban environments
Objective 5:
• Increased market share and revenue of European companies through PRYSTINE’s ground¬breaking technological advancements (O1-O4)
Objective 6:
• Increased user acceptance of automated driving functions through PRYSTINE’s groundbreaking technological advancements (O1-O4)
Moreover, PRYSTINE has followed a very concrete dissemination, communication and exploitation strategy, that has resulted in the following:
a) 30 project posters
b) 4 leaflets/flyers
c) 6 e-newsletters
d) 20 open-aire publications
e) 105 journal and conference publications
f) participation in total in 81 events
g) ~100 projects exploitable results were extracted from the questionnaires and can be categorized in three major categories according to its type: Knowledge or know-how, software and hardware components, and services or methodologies
The progress beyond the state of the art (SOTA) can be summarized on a per objective basis, as follows:
1. Progress has consisted in the development the core and fundamental building bricks / components of the higher-level PRYSTINE demonstrators.
2. Progress beyond the SOTA lied in pushing the technological capabilities for autonomous driving upwards (SAE Level 3+) by implementing in PRYSTINE fail-operational mechanisms for embedded control and enhancing sensor fusion by cameras, Lidar, Radar. This objective is realized on the system level and is fulfilled through the implementation and validation of the following demonstrators.
3. Work here has provided an environment for the development an optimised fail-operational system architecture to facilitate the progress of driving automation functions from the standard fail silent ADAS features. The architecture considers the integration of a full range of components (sensors, controllers, actuators and communication hardware) necessary to enable AD, while also taking into account the dependability aspect.
4. The following areas have been progressed beyond the state of the art:
- A sensor fusion S/W block has been integrated to receive input from the perception block and generate outputs to map generator and path planning blocks on HDT‘s target platforms.
- Development of multi lidar sensor fusion, path planning, motion control and environmental awareness block which can slowly brake and stop the car unless the obstacle is in the LiDAR coverage area.
- Development of sensor fusion algorithm and construction gate detection system.
- Integration of the back-manoeuvring assist solution on heavy duty vehicles parking at docking station and construction sites.
- A heavy-duty truck has been brought to test ready status with computing platforms and perception sensors. UTU has developed the software part of Demo 5.2 and integrated in the ABOX computing platform of the TTS heavy-duty truck.
- Traffic Light Time-to Green
- Trajectory Recognition and VRU
- Emergency Lateral Lane Stop
- Shared Control (focuses on the development of an intelligent co-driver with adaptive level of assistance based on driver-automation status and different risk indicators that are present during the driving task).
- Traded control (real automated vehicle in a closed driving environment)
- AI based Decision Making
5. Provision of competitive advantages for the semiconductor Industry and knowledge institutions. It includes market impact and novel market figures.
6. The main challenge here has been twofold: how to increase the user acceptance of automated driving functions (ADFs) and how to strength the competitive advantage for European industry.
The progress beyond the SOTA has been focused on the measurements of users’ expectations and trust in ADFs as developed by PRYSTINE, as well as the assessment of the effectiveness of adaptation, drivers’ situation awareness and trust in automation.
1. Progress has consisted in the development the core and fundamental building bricks / components of the higher-level PRYSTINE demonstrators.
2. Progress beyond the SOTA lied in pushing the technological capabilities for autonomous driving upwards (SAE Level 3+) by implementing in PRYSTINE fail-operational mechanisms for embedded control and enhancing sensor fusion by cameras, Lidar, Radar. This objective is realized on the system level and is fulfilled through the implementation and validation of the following demonstrators.
3. Work here has provided an environment for the development an optimised fail-operational system architecture to facilitate the progress of driving automation functions from the standard fail silent ADAS features. The architecture considers the integration of a full range of components (sensors, controllers, actuators and communication hardware) necessary to enable AD, while also taking into account the dependability aspect.
4. The following areas have been progressed beyond the state of the art:
- A sensor fusion S/W block has been integrated to receive input from the perception block and generate outputs to map generator and path planning blocks on HDT‘s target platforms.
- Development of multi lidar sensor fusion, path planning, motion control and environmental awareness block which can slowly brake and stop the car unless the obstacle is in the LiDAR coverage area.
- Development of sensor fusion algorithm and construction gate detection system.
- Integration of the back-manoeuvring assist solution on heavy duty vehicles parking at docking station and construction sites.
- A heavy-duty truck has been brought to test ready status with computing platforms and perception sensors. UTU has developed the software part of Demo 5.2 and integrated in the ABOX computing platform of the TTS heavy-duty truck.
- Traffic Light Time-to Green
- Trajectory Recognition and VRU
- Emergency Lateral Lane Stop
- Shared Control (focuses on the development of an intelligent co-driver with adaptive level of assistance based on driver-automation status and different risk indicators that are present during the driving task).
- Traded control (real automated vehicle in a closed driving environment)
- AI based Decision Making
5. Provision of competitive advantages for the semiconductor Industry and knowledge institutions. It includes market impact and novel market figures.
6. The main challenge here has been twofold: how to increase the user acceptance of automated driving functions (ADFs) and how to strength the competitive advantage for European industry.
The progress beyond the SOTA has been focused on the measurements of users’ expectations and trust in ADFs as developed by PRYSTINE, as well as the assessment of the effectiveness of adaptation, drivers’ situation awareness and trust in automation.