Research On Algorithms Of The Shortest Path For Autonoumous Vehicle Under Multiple Constraint Condition

As one of the core technologies of intelligent driving,path planning has always been a research hotspot in the field of unmanned vehicles.The shortest path planning is undoubtedly the focus of the research direction of path planning because it can fully reduce the travel cost of the overall path and increase the utilization rate of the driving space.Moreover,the shortest path planning is an effective way to solve traffic congestion,improve traffic efficiency,and reduce overall vehicle energy consumption.In addition,the passage of unmanned vehicles is also subject to various traffic scenarios and driving conditions,such as collision-free constraints,multi-node constraints,speed smoothing constraints,etc.,to ensure the safety,efficiency,and stability of unmanned vehicles.Therefore,studying the shortest path planning under multiple constraints has important theoretical value and practical significance.The main research contents and innovations are as follows:(1)Shortest path planning without collision constraintsThe shortest path planning under collision-free constraints introduces lidar sensors for the construction of collision-free constraints.The cost function design for collision-free constraints is designed to ensure that the path points to be selected meet the collision-free constraints.At the same time,a collision-free constraint is proposed.Constrained A*algorithm,the path points selected by the algorithm need to meet the cost function of the collision-free constraint and the shortest path cost function at the same time,in order to plan the shortest path,and ensure that each step of the algorithm search meets the collision-free constraint and the shortest path It is required to finally plan a shortest path for unmanned vehicles without collision constraints.(2)Shortest path planning under node constraintsThe node-constrained shortest path planning first defines the conditions of the node constraints.It is clear that the algorithm needs to traverse those nodes in addition to the starting point and the end point,and then constructs the shortest path cost function under the node constraints according to the node constraints,and then integrates the cost function into The Dijkstra algorithm uses the nature of Dijkstra’s greedy strategy to gradually accumulate the global shortest path under the condition of meeting the node constraints,until the algorithm explores the end point,and finally generates the shortest path under the node constraints.(3)Shortest path planning under speed constraintsFirst,the shortest path planning under the speed constraint uses the third-order B-spline curve to construct the speed constraint conditions based on the derivable characteristics of the third-order B-spline curve,so as to select the curve control points.After the control points are determined,A unique curve of path based on the current control point can be obtained.For the shortest path planning,it is only necessary to include the starting point and the end point in the selection of the control point,so that the obtained path is the shortest path under the current speed constraint,and it is also the only one path.