Increasing electronic and software content has resulted in exponentially growing complexity of modern mechatronic systems such as automobiles, aircrafts, railway systems. This is particularly true for connected autonomous vehicles, equipped with a variety of sensors, and intelligence spanning all the levels of autonomy. Such systems typically consist of software and hardware components that are developed independently and reused through integration. A critical consequence is the emergent uncertainty about the integrity and performance of the final deployed system. Such uncertainty is safety critical for systems such as the autonomous vehicles, where there is acute interaction between humans and machines.
Verification and Validation (V&V) of such systems against the original requirements is imperative for providing the needed confidence/trust in such complex cyber-physical systems. Performing the V&V tests only on the final end-product is prohibitively expensive and time consuming. The modern approach to V&V is to use computer models of cyber physical systems that can be simulated virtually.
This Virtual Autonomous Vehicle Simulation (VAVS) TDP is intended to be platform that can be used for simulating an individual vehicle or a fleet of vehicles, manually or autonomously driven, in varied traffic situations and road conditions.
The VAVS-TDP will be an integral part of most research projects that will be executed as part of the CAST Program. The VAVS-TDP will be used for control development, design optimization, preliminary calibration, and most importantly verification of autonomous vehicle systems. The VAVS-TDP will also become a key enabler for research on human-machine-interactions (HMI).
The VAVS-TDP will also be leveraged to support “industry best practices” for development of control systems. In particular, the VAVS-TDP will be an important platform for practicing and teaching Model Based Systems Engineering (MBSE) – including development of requirements, architecture, etc.
The development of the VAVS-TDP itself is expected to uncover new research challenges. There are multiple dimensions of challenges such as in the scale of simulation, speed of simulation (especially real-time simulation), modeling and simulation of new sensors, communications, human machine interaction simulators (including Virtual Reality and Augmented Reality), etc. These challenges will be addressed through individual research projects.
- Modeling and Simulation
- Individual and Multiple vehicle simulations of longitudinal and lateral dynamics
- MIL, SIL and HIL capabilities
- MBSE Processes
- Requirements capture and analysis
- Architecture capture and analysis
- Traceability between requirements and testing
Projects Leveraging the TDP
- Safe-Core Architecture We would like to explore and develop a reference “Safe-Core Architecture” that becomes the basis for developing the AV Embedded Control Software
- Middleware to connect Robotics and Automotive architectures for Autonomous Vehicle Control We would like to explore architectures and implementation methodologies that would result in efficient integration of ROS and traditional RTOS based control algorithms.