SMARTER AI FOR SAFER TRANSPORTATION
Claude Hochreutiner
Edge AI can benefit fleets to be used to rate drivers and identify whether they meet the necessary safety standards
A report by the Stanford Law School estimates that at least 90 percent of crashes are at least partially due to driver error. Additionally, a study by the Insurance Information Institute in 2019 found that driving recklessly without regard for following traffic rules was the primary cause of accidents.
Specifically, the study lists driver behavior-related reasons for accidents in the United States as follows:
Reason
Percentage of Accidents
Driving recklessly and disregarding traffic rules: 28.1
Speeding: 17.2
Driving under the influence of alcohol, drugs or medication: 10.1
Driving while distracted (using the phone, talking, eating): 5.9
Driving while drowsy, fatigued or ill: 2.4
From this, it follows that monitoring drivers and warning them about reckless behavior has a considerable impact on road safety.
Edge AI in Telematics
Drivers must be monitored and warned in real-time so that accident-causing behavior can be modified early enough to prevent accidents. The need for quick responses means that the telematics devices that monitor (and warn) vehicle drivers must process data locally. Companies produce smart dashcams that are specifically tailored for such scenarios. These telematics devices use AI to process the inputs they generate to produce instantaneous outputs. The shifting of AI-related processing from a centralized location, such as a cloud, to the AI chip embedded in the device, is called AI on the Edge or Edge AI. Because data need not be uploaded on the cloud to be processed, Edge AI:
Reduces the time required to process an event.
Eliminates the costs to send and receive data.
Eliminates the need to have continuous connectivity to process the data.
Reducing human error with Edge AI
Using Edge AI can help prevent accidents by:
Warning drivers about rash driving — Monitoring drivers for rash behavior and issuing warnings in real-time gives the driver a chance to proactively correct such behavior. In fleets where there are multiple drivers, this system can also be used to rate drivers and identify whether they meet the necessary safety standards.
Warning drivers about potential hazards — Identifying potential hazards in the path of the vehicle and warning drivers in real-time allows the driver enough time to take appropriate action. In certain cases, the system can automatically brake to prevent the vehicle from colliding.
Monitoring drivers for drowsiness — AI systems can be taught to recognize telltale signs of a drowsy driver, such as excessive blinking or yawning, and issue a warning when these signs are detected. In some cases, the system can take action. For example, reduce the ambient temperature, to prevent the driver from falling asleep.
Warning drivers about distracted driving — Detecting and warning the driver about distracted driving behavior, such as using cell phones or eating while driving, will highlight such behavior and encourage the driver to pay attention.
In conclusion, it is up to us to decide how safe our roads are. Human error is one of the leading causes of accidents and minimizing it could lead to fewer road accidents. One way to do this is to use devices with Edge AI, which can identify potential hazards in real-time and instantaneously issue warnings and take action. They are always running and never tired, with the added advantage of being able to act, and can help make our roads safer.
Claude Hochreutiner has been active in the security, IT and IoT industry for more than 15 years, and has worked for General Electric, Thales, Schneider Electric and Bosch in Europe, the Middle East and Asia Pacific. Hochreutiner holds a master’s degree in Communications Systems from the Swiss Federal Institute of Technology (EPFL) as well as a PMP certification.
https://www.motor.com/2022/02/smarter-ai-for-safer-transportation/
