AI Can Save Lives 

Artificial intelligence is becoming a key driver for greater safety and fewer accidents on the roads. Continental is at the forefront of the development of such systems – and is thus getting ever closer to achieving Vision Zero. Its ambitious goal: a world without road accident victims.  

Road traffic has become safer – but it is still a long way from being safe enough. Airbags, seatbelts, ABS and ESC, as well as legal regulations such as mandatory seatbelt use, have all made a significant contribution to reducing the number of accidents involving serious injuries and fatalities. Continental was and still is a leader in the development of such road safety technologies. Back in 1974, more than 20,000 people died in road accidents each year in Germany, despite there being significantly fewer cars on the roads at the time. Today’s figure sits at around 3,000. This signifies great progress, but the number is still far too high.    

There is a new driver for greater safety and fewer accident victims, however: artificial intelligence. Smart software is becoming a decisive factor for Continental’s Vision Zero. For a world without road accident victims. Thanks to software applications in new advanced driver assistance systems that learn from increasing amounts of data, the number of accidents will continue to decrease – and fewer people will be injured on the roads. The environmental perception of sensor systems and the processing of large amounts of data in real time have now reached a level of maturity that surpasses human capabilities.  

If we take a look at the electronic architecture of a modern vehicle, this becomes clear: AI applications are already integrated into systems for emergency braking, lane departure warning, blind spot detection, right turns and traffic sign recognition – and they will continue to have an impact in future generations of assistance systems. This is backed up by independent studies. The European Road Safety Council, for example, assumes that advanced driver assistance systems will be able to reduce the number of road fatalities by up to 30 percent thanks to their use of AI. A study by the US Insurance Institute for Highway Safety (IIHS) found that vehicles equipped with a collision avoidance system and an emergency braking system reduce rear-end collisions by 50 percent.  

But that is just the start. There is great potential in monitoring the driver’s level of attention in particular. The complexity of traffic, health problems or the driver momentarily nodding off can all be recognized by intelligent systems, and accidents can be avoided as a result. According to the US National Highway Traffic Safety Administration (NHTSA), 94 percent of all traffic accidents are caused by driver errors. Drowsiness detection systems alone can reduce the risk of accidents caused by inattentiveness at the wheel by up to 40 percent. According to experts, innovative functions such as personalized feedback from AI-based sensor systems will lead to a 20 percent reduction in the occurrence of a risky driving style.    

The use of AI in safety systems is still in its early stages. The faster the degree of automated driving functions in cars increases, the safer mobility becomes. A look at the specific causes of serious accidents in Germany due to driver errors in 2022 makes this clear. Errors when turning, during drive-off or when entering a road are the main culprits at 15.3 percent, followed by disregarding the right of way (13.4 percent), failing to maintain a safe distance (12.9 percent) and inappropriate speed (11.4 percent). Most of the causes can easily be avoided by means of automated driving.  

An AI system is only as good as the data it is trained with, however. The number of possible traffic situations is almost unmanageable, and the probabilities of certain situations occurring can be very low. Continental is therefore taking different approaches to how it designs AI technology and is already using AI in the development and testing of systems. Generative AI – software that independently generates new content and therefore datasets – can therefore be a powerful tool for creating test data for autonomous driving systems. From the generation of even unusual environmental variants and traffic patterns to very unlikely accident scenarios – so-called “edge cases” – data can now be generated using generative AI models. This allows test datasets to be significantly improved – and with this the effectiveness of AI-based assistance systems as well.    

The final step towards achieving Vision Zero through autonomous driving is yet to come. When cars are able to drive themselves in the future, and the “human error source” in the vehicle no longer plays a role in principle, AI systems will then be able to realize their full potential for greater safety in road traffic. Then the goal of a world without accidents – Continental’s Vision Zero – really will be close.  

• In a webinar titled “The Future of Vehicle Safety – Can AI Enable Vision Zero?” our experts provided insights into the history and current role of AI at Continental with a specific focus on our Vision Zero. This webinar offers a glimpse into our exciting AI initiatives at Continental. Click here for the recording: The future of vehicle safety – can AI enable Vision Zero?