Managing Editor of Future Vehicle magazine, Steve Welch, interviewed Mike Dempsey, Managing Director of Claytex prior to the virtual LCV2020 exhibition.
Simulation will play a key role in proving the reliability and trustworthiness for testing autonomous vehicles and advanced driver-assistance systems.
The commercial development and market adoption of autonomous vehicles will rely, in no small part, on confidence in the safety of such driverless systems. It seems certain that digital modelling and simulation tools will, alongside more conventional physical testing, play a key role in proving the reliability and trustworthiness of advanced driver-assistance systems (ADAS) and fully autonomous platforms.
The complexity of simulating ADAS (let alone fully autonomous vehicles) requires processing power beyond the typical capabilities of a desktop machine. In response, the automotive simulation and modelling company Claytex is now operating the driving simulation software rFpro distributed across several machines. “You can’t really carry on with the type of simulation that has been used for the past ten years: a single simulation programme simulating one car,” comments Claytex managing director Mike Dempsey. “It’s just not enough when you have a full suite of sensors that all need to be synchronized; and you really need to be using a physics-based approach for everything.”
Autonomous vehicles could have some 40 perception sensors running concurrently, including multiple cameras, radar, LiDAR, ultrasound and GPS. According to Dempsey, there is little value in looking for short cuts as you have to have confidence that the system will not fail in any real-world context. “You need to be able to simulate all of the sensors, and if you drop the detail level in any one of those, it makes it too easy for the perception algorithms to detect what’s happening,” he says. “So, with all of those sensors you need to have a pretty high level of detail in that simulation to be able to do a good job of simulating what that sensor is going to be able to see, detect, and feed in to the perception algorithm.”
Read the full article here
Please feel free to contact email@example.com to get more information on our capabilities and our range of systems engineering solutions.
Telephone: +44 1926 885900