Team of engineers at the University of Waterloo have created a new system which focuses on minimizing injuries and damage in self-driving vehicles.
The new decision-making and motion-planning technology can recognize an inevitable collision. By figuring out all the possible available options, the system helps in choosing an action with the least serious outcome.
One of the primary rules of autonomous vehicle (AV) crash-mitigation technology is to avoid collision with pedestrians. It is followed by other factors such as angles of collision, relative speeds, vehicle types and differences in mass. This further helps in determining the best possible manoeuvre, which may include steering in another direction or breaking.
Experts who are focusing on minimizing the consequences of AV say they had to consider the entire traffic environment which is perceived by AV, which also includes other vehicles and obstacles.
Furthermore, the experts highlight the need of such a system because it is only a myth that AVs in the future will eliminate crashes. According to Amir Khajepour, a professor of mechanical and mechatronics engineering at the University of Waterloo, although safety could improve in the future, he draws attention to the several uncertainties of AVs to be handled without the possibility of any mishaps.
Furthermore, AVs have the capability to limit damage as they are more aware of their environment via sensors, cameras, in addition to other sources. Experts allege, AVs can make hundreds of decisions per second. The new system will enhance the response of an AV in case of an emergency. This are made possible through pre-defined mathematical calculations after considering the intensity of the damage and crash injuries.