Cattle grid or wall? ADAS and the simulation opportunity
02 August 2021
It is clear that autonomous driving is part of our transport future, and as motorists around the world relinquish even more control to vehicles fitted with advanced driving assistance systems (ADAS), trust in a vehicle has never been more important.
Engineers are striving to develop, test and verify vehicles to intertwine human and machine, in order to create and ensure trustworthy ADAS systems. However, driving can often be unpredictable. Every day, drivers contend with varying weather conditions, pedestrians, cyclists, roadworks, and animals (to name a few). So, what happens if the vehicle cannot correctly predict or interpret the road in front of it?
To date, there have been some extreme examples of failings in autonomous vehicle technology. In 2018, during some of the first tests of self-driving vehicles, reports appeared in the media of fatal crashes, with vehicles colliding with parked police cars and crashing into cyclists. In response, vehicle development professionals have developed far more rigorous testing and simulation to help predict these things.
A more recent issue in Somerset, UK has again sparked concern. ADAS-equipped vehicles on a particular stretch of road were mistaking a cattle grid for a wall, which resulted in a number of vehicles’ ADAS systems either applying the brakes or veering off the road. The results were damaged vehicles, the drivers’ shaken trust in the capabilities of ADAS systems, and, ultimately, a redesign of the cattle grid, costing the local council thousands to fix.
For the vehicle development community, this raises some serious questions: namely, how can this be prevented from happening again? What measures can be put in place to guarantee the safety of advanced, technology-driven vehicles on the road?
The answer is simulation.
Read the full article in the August issue of DPA
Contact Details and Archive...