In order to predict the passing ability of autonomous vehicles on structured roads

a collaborative control method for vehicle obstacle avoidance and passing ability is established. Aiming at small and medium-sized obstacles or trenches that affect vehicle passing ability

a joint calibration and information acquisition system using millimeter-wave radar and camera is created.The geometric bounding box model parameters of the target obstacle are extracted

and the data are matched and fused using bounding box feature values to form high-reliability obstacle feature information.A mechanical model for vehicle passing ability is established

and the driving limit data that the vehicle can pass through are derived.According to the position and characteristic information of obstacles

combined with the passing ability limit parameters of vehicle

the passing ability of autonomous vehicle is predicted and reasonable traffic and obstacle avoidance control strategies are proposed.Vehicle obstacle avoidance and passing ability tests are conducted to verify the feasibility of vehicles passing through vertical obstacles and crossing trenches

and improve the safety of intelligent vehicle driving.