A German research team has developed a safety system that can warn drivers of self-driving cars up to seven seconds in advance that they will need to take control. A team of researchers at the Technical University of Munich (TUM) has created a new early warning system for self-driving cars that employs artificial intelligence to learn from thousands of real-world traffic situations. The system was researched in collaboration with the BMW Group.
The study’s authors claim that if used in today’s self-driving cars, it could provide seven seconds of advanced warning against potentially critical situations that the cars cannot handle on their own – with over 85% accuracy.
To make self-driving cars safer in the future, development efforts frequently rely on sophisticated models designed to give cars the ability to analyze the behavior of other traffic. However, researchers are now wondering what will happen if the cars encounter a complex situation they have never seen before and are unable to handle on their own.
A team of researchers in Germany has come up with a safety system that could warn drivers of autonomous cars that they will have to take control up to seven seconds in advance.
A team led by Prof. Eckehard Steinbach, Chair of Media Technology and member of the Board of Directors of TUM’s Munich School of Robotics and Machine Intelligence (MSRM), is taking a novel approach.
They claim that artificial intelligence (AI) allows their system to learn from previous situations in which self-driving test vehicles were pushed to their limits in real-world traffic. These are situations in which a human driver takes over – either because the car indicates that intervention is required or because the driver decides to intervene for safety reasons.
The system was studied in collaboration with the BMW Group, and the results were published in the journal IEEE Transactions on Intelligent Transportation System. The results showed that, when used in today’s self-driving cars, it can warn seven seconds ahead of potentially critical situations that the cars cannot handle on their own – with over 85% accuracy.
Pattern recognition through RNN
Sensors and cameras are used to capture surrounding conditions and record vehicle status data such as steering wheel angle, road conditions, weather, visibility, and speed. Based on the data, the AI system, which is based on a recurrent neural network (RNN), learns to recognize patterns. If the system detects a pattern in a new driving situation that the control system was unable to handle previously, the driver will be warned of a potentially dangerous situation.
“To make vehicles more autonomous, many existing methods investigate what cars currently understand about traffic and then attempt to improve the models that they use. Our technology has the significant advantage of allowing us to completely disregard what the car thinks. Instead, we limit ourselves to data based on what actually occurs and look for patterns,” Steinbach explained.
“In this way, the AI identifies potentially critical situations that models may be incapable of recognizing or have yet to discover. As a result, our system includes a safety function that detects when and where the cars have flaws.”
Warnings up to seven seconds in advance
The researchers tested the technology on public roads with the BMW Group and its autonomous development vehicles, analyzing approximately 2500 situations in which the driver had to intervene. According to the research, AI is already capable of predicting potentially critical situations with greater than 85% accuracy – up to seven seconds before they occur.
Large amounts of data are required for the technology to function. According to the research team, AI can only recognize and predict experiences at the system’s limits if the situations have been seen before. According to Christopher Kuhn, one of the study’s authors, the data was practically generated by itself due to the large number of development vehicles on the road.
Every time a potentially critical situation arises during a test drive, we create a new training example,” he explained. According to the researchers, central data storage allows each vehicle to learn from all of the data collected across the entire fleet.
