Digitization of Traffic Laws: Methodologies and Usage for Monitoring Driving Compliance

Autonomous Vehicles (AVs) must adhere to traffic laws designed for human-driven vehicles. However, since current traffic laws are expressed in natural language and are inherently ambiguous, AVs encounter challenges in comprehending these laws. Therefore, digitizing traffic laws into a format that AVs can understand is crucial for safe and efficient driving. In this paper, a process for digitizing regulations is proposed, where each regulation is digitized into a temporal logic expression composed of computable atomic propositions. Based on this process, a vehicle-side online violation monitoring architecture is established. These works make AVs understand traffic laws easily. Several common but important regulations are used as examples to illustrate our work and are deployed on an AV for verification. The results demonstrate that the proposed monitoring architecture can monitor ego vehicle's illegal behavior in real-time and provide compliance suggestions, thereby helping AVs operate more safely.