A Response-Type Road Anomaly Detection and Evaluation Method for Steady Driving of Automated Vehicles

Serious Road anomalies caused by bridge approach settlement, pavement rutting, etc., not only seriously affect traffic safety and user experience but also aggravate the damage of road structure. It is more relevant to automated vehicles (AVs) as they are currently not designed to measure the pavement roughness directly. Without prior-collected road anomalies information, AVs' active suspension control system can only passively reduce the negative impact of road anomalies to a certain extent. This paper proposed a response-type road anomaly detection and evaluation method by collecting the vibration data from AVs. A mechanical estimation model for the height of anomaly (HoA) is constructed to evaluate the degree of road anomaly. Passenger's comfort is evaluated by three featured indicators: maximal acceleration, weighted root-mean-square acceleration, and jerk. A full-car simulation model is programmed based on the Simulink platform to reveal the relationship among road anomalies, comfort, and speed, which helps design a steady driving velocity profile for AVs. The results show that the root-mean-square error of road anomalies estimation is about 0.63cm. AVs' comfort can be improved significantly by employing the proposed steady driving strategies.