| With the gradual deepening of the intelligent information age,industrial empowerment is becoming increasingly important,and the transformation of technology and intelligence is an important direction for industrial development.The important transformation of the automotive industry is that vehicles are moving towards low pollution and intelligence,which can largely solve safety problems caused by manual or artificial factors and alleviate the symptoms of urban transportation inconvenience.In the parking lot environment,there are many vehicles and pedestrians,and the environmental structure is complex,so the implementation of vehicle intelligence is more difficult.As a core part of the autonomous driving framework,path planning also needs to face enormous challenges.This article proposes a path planning algorithm suitable for complex working conditions in parking lots.This algorithm achieved good results during the testing process and was equally stable in the actual map planning process.This article summarizes and analyzes the current research status of path planning algorithms for intelligent vehicles and robots at home and abroad,and combines the current development status of intelligent vehicles.It is found that the efficiency of path planning algorithms cannot be guaranteed in complex environments,especially in complex working conditions such as parking lots.Therefore,research is conducted on path planning algorithms for intelligent vehicles in parking lot environments.The specific content is described as follows:Firstly,in terms of path planning algorithm and modeling of intelligent vehicles,the concept and classification of path planning algorithm are studied,and the basic principles and processes of common path planning algorithm,environment model and vehicle model are introduced in detail respectively.The advantages and disadvantages of the algorithms are compared and analyzed,which provides support for the direction of the subsequent improvement algorithm.Secondly,in the aspect of global path planning of intelligent vehicles,in order to ensure that intelligent vehicles can safely travel according to the planned path under the condition that they meet the vehicle dynamics characteristics,a new algorithm based on Probabilistic and hybrid A * algorithm-probabilistic hybrid A * algorithm is studied and proposed to solve many unnecessary search problems caused by the lack of guidance strategies in hybrid A * algorithm.Firstly,the Probabilistic A *algorithm is proposed to obtain a series of guidance points,and then the obtained guidance points are used to guide the node direction of the hybrid A * algorithm,avoiding nodes from searching towards obstacles.Finally,optimize the cost function of the node.Through comparative experimental analysis with A * algorithm and existing hybrid A * algorithm and their improved algorithms,the path obtained is relatively regular and smooth,thus verifying the effectiveness and feasibility of the Probabilistic hybrid A * algorithm.Thirdly,in terms of local path planning for intelligent vehicles,in order to adapt to the global path planning of intelligent vehicles and combine the characteristics of global path planning,an artificial potential field algorithm based on Probabilistic A *algorithm-Probabilistic artificial potential field algorithm was studied.The algorithm combines the guidance points generated by the Probabilistic A * algorithm with the artificial potential field method,solving the problem of local minimum points and unreachable targets in the search process of the artificial potential field algorithm.Simultaneously adding dynamic factors and optimizing relevant parameters.The effectiveness and feasibility of the artificial potential field algorithm based on probability A * algorithm were verified through simulation experiments.Finally,the vehicle global path planning algorithm and local path planning algorithm proposed in this article are evaluated.Through simulation experiments,compared with traditional algorithms,the proposed algorithm reduces planning time by an average of 13.5% and the search path length by an average of about 7.1%.Future improvement directions are also proposed. |
