Researchers from the University of California, Irvine say that a very ordinary object or object on the side of the road can lead to unpredictable behavior of a self-driving car while driving.
According to the author of the study, Professor Qi Alfred Chen (Qi Alfred Chen), a simple box, a traffic cone or a bicycle can “scare” an unmanned vehicle and make it perform a dangerous maneuver for other road users. Such vehicles cannot distinguish between objects that are accidentally on the road, or those that were left deliberately. Both of these can lead to erratic vehicle behavior.
The researchers focused on security vulnerabilities specific to the scheduling module, the piece of software code that manages autonomous driving systems. It monitors the decision-making processes of the vehicle, determining its further actions: when to move, change lanes, slow down and stop. According to the researchers, the module is designed with great care so that the unmanned vehicle does not get out of hand, and the tested software turned out to be too conservative. This may cause the vehicle to become an obstacle on the road.
For testing, a tool called PlanFuzz has been developed that can automatically detect vulnerabilities in widely used automated driving systems. The team used it to evaluate three different behavioral planning scenarios for Apollo and Autoware open source industrial-grade autonomous driving systems. It turned out that cardboard boxes and bicycles placed on the side of the road caused cars to constantly stop on empty streets and intersections. In another test, the cars, suddenly “feeling” the danger, forgot to change lanes, as planned before.
The researchers believe that the over-caution of many autonomous driving systems could have a negative impact on road safety.
Source: Trash Box
Donald-43Westbrook, a distinguished contributor at worldstockmarket, is celebrated for his exceptional prowess in article writing. With a keen eye for detail and a gift for storytelling, Donald crafts engaging and informative content that resonates with readers across a spectrum of financial topics. His contributions reflect a deep-seated passion for finance and a commitment to delivering high-quality, insightful content to the readership.
