According to the Technical University of Munich (TUM), the latest software developed by one of its departments is capable of reassessing changing traffic situations millisecond by millisecond.
The institution says the software has been conceived to handle very complex traffic interactions, such as those at intersections where pedestrians, other vehicles (which can act erratically) and weather conditions all come into play.
“These kinds of situations present an enormous challenge for autonomous vehicles controlled by computer programs,” explained Matthias Althoff, Professor of Cyber-Physical Systems at TUM. “But autonomous driving will only gain acceptance of the general public if you can ensure that the vehicles will not endanger other road users – no matter how confusing the traffic situation.”
He noted that the ultimate goal when developing software for autonomous vehicles is to ensure that they will not cause accidents. Althoff, who is a member of the Munich School of Robotics and Machine Intelligence at TUM, said that he and his team have now developed a software module that permanently analyzes and predicts events while driving. Using high-fidelity vehicle sensor data, the software calculates all possible movements for every traffic participant – provided they adhere to the road traffic regulations – allowing the system to look three to six seconds into the future.
Based on these future scenarios, the system determines a variety of movement options for the vehicle. At the same time, the program calculates potential emergency maneuvers in which the vehicle can be moved out of harm’s way by accelerating or braking without endangering others. Constraints within the programming mean the autonomous vehicle may only follow routes that are free of foreseeable collisions and for which an emergency maneuver option has been identified.
According to Althoff, this type of detailed traffic situation forecasting was previously considered too time-consuming and thus impractical. But now, the Munich research team claims to have shown not only the theoretical viability of real-time data analysis with simultaneous simulation of future traffic events, but also demonstrated that it delivers reliable results.
The quick calculations rely on simplified dynamic models. So-called reachability analysis is used to calculate the potential future positions a car or a pedestrian might assume. When all characteristics of the road users are taken into account, the calculations become prohibitively time-consuming, hence the need for simplified models. The team states these are superior to the real ones in terms of their range of motion – yet, mathematically easier to handle. This enhanced freedom of movement allows the models to depict a larger number of possible positions but includes the subset of positions expected for actual road users.
For their evaluation, Althoff’s team created a virtual model based on real data they had collected during test drives with an autonomous vehicle in Munich. This allowed them to craft a test environment that closely reflects everyday traffic scenarios.
“Using the simulations, we were able to establish that the safety module does not lead to any loss of performance in terms of driving behavior, the predictive calculations are correct, accidents are prevented, and in emergency situations, the vehicle is demonstrably brought to a safe stop,” Althoff summed up.