Abstract

Detecting geometric road anomalies in digital maps at a large scale can present a challenge, as some conventional methods may exhibit limitations in precision and result in a higher rate of false positives. A method and system are described that can identify these anomalies by simulating a vehicle's traversal of a given road segment to assess its kinematic feasibility. The model can analyze the change in the road centerline's bending energy before and after lane rendering and can predict the probability of a vehicle spin-out based on lateral acceleration. This approach may allow for accurate, scalable detection and quantification of road geometry errors, thereby contributing to the quality and reliability of digital maps for navigation.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.

Share

COinS