Motivation

Automated vehicles will be used in many areas in the future, such as passenger transportation, mining, agriculture, container transport, and many more. In designated areas, automated vehicles are already used today, e.g. the port of Hamburg uses unmanned vehicles for container transportation. Especially in the field of passenger transportation, the research activities have recently increased dramatically due to a number of advantages:

  • Collision avoidance: Worldwide around 1.2 million people are killed in road accidents (according to the world health organization). Future automated vehicles can predict road scenes and plan their actions such that under no circumstances a crash is caused.
  • Improved throughput: Automated vehicles improve the throughput on roads by driving closely together (platooning). They can also coordinate actions around traffic intersections such that no vehicle has to wait, but only adjust the speed to cross an intersection.
  • Mobility for children and elderly: People who are not capable of driving can use automated vehicles for personal mobility.
  • Comfort: People who daily commute to work can use the traveling time to do work or relax.
  • Valet parking and car sharing: One can always order a car to any position, such that mobility is provided anywhere. The vehicle would drive itself to the place where the passenger likes to be fetched from.

Objective

In order to use automated vehicles in unpredictable environments, such as road traffic, the safety of unmanned vehicles should be formally verified. The objective is to develop methods that can guarantee the safety of autonomous vehicles online, i.e., while the vehicle is in operation. This is required since the safety specification depends on the current scene. The algorithm has to be conservative in the sense that even in the presence of unforeseeable events, safety is guaranteed.

Development

Currently, we are testing our formal approaches on real vehicles and continue to develop out testing platform CommonRoad.