Research On The Mutli-intelligent Vehicle Collaborative Control Based On Artificial Potential Field

In order to improve the traffic efficiency and safety of intelligent vehicle,this article takes multi-intelligent vehicle collaborative control as the research object,discusses and analyzes the formation control of multi-intelligent vehicles under non-road conditions and road conditions and the path planning of single vehicle and formation obstacle avoidance based on the artificial potential field.To realize the formation control behaviour such as formation formation,formation changes and formation maintenance based on a multi-agent formation control model,which combined the artificial potential field and virtual leader.The new goal potential field model and the formation unit were established and the longitudinal and horizontal iteration of unit was set based on the decomposition-iterative method,which improves model application flexibility and the position error of formation was eliminated through controlling the position accuracy of vehicle in the unit.Then,the fixed-point potential field and the "single road" boundary potential field were introduced and the concept of "flexible" virtual leader radius of action was proposed based on the above work,which ensures the accuracy of the position of the formation vehicle and realizes the formation change behavior under non-road conditions.Then,the formation ideal behavior in highway was analyzed and the minimum longitudinal and horizontal safety distance model of formation and the formation security constraints model were developed.Based on the models characteristic,a new virtual force scope of virtual leader was proposed and the adjustment factors were introduced to change the size and orientation of the new virtual leader’s potential field.The simulation result of two-lane and six-vehicles formation shows that the formation model can safely and effectively realize the formation of multi-intelligent vehicles in the hybrid highway.For the study of vehicle obstacle avoidance path planning,firstly,the shortest longitudinal safety distance and overtaking constraint model of overtaking were established through analyzing the law of overtaking driving behavior.Based on artificial potential field and the overtaking model,a new virtual force scope of obstacle was proposed.The repulsive force coefficient based on the front vehicle operating conditions and a virtual local goal point theory was built,thus an improved path planning model is established.The path tracking simulation result,which based on Model Predictive Control,shows that the improved path planning mode can safely and effectively make the intelligence vehicle achieve overtaking behavior.Finally,according to the relative position of obstacles and formation vehicles,the different level of danger areas were divided.And the different obstacle avoidance forms of vehicles in different danger level areas were discussed and analyzed and the methods of formation direct obstacle avoidance was proposed based on the research of the above single vehicle obstacle avoidance.The simulation results of formation avoidance path show that the proposed method of formation avoidance can safely and effectively realize the formation obstacle avoidance behavior under the premise of the formation remain in shape.The above research results can be widely applied to intelligent vehicle control engineering,especially in the practical application of intelligent vehicles.It will make the high academic value and broad application prospects.