Research On Path Planning For Off-Road Unmanned Target Vehicles

With the continuous advancement of technology,artificial intelligence(AI)has become one of the hot topics and trends in today’s society.As a typical representative of AI technology,autonomous vehicles are at the forefront of research and development in the automotive industry.Among them,unmanned target vehicles,as a special type of autonomous vehicle,have broad application prospects,mainly in military,scientific research,and testing fields.Compared to traditional driving vehicles,unmanned target vehicles operate in relatively complex scenarios.Instead of focusing on comfort,they prioritize the feasibility of paths and the safety of vehicle operation.Therefore,in off-road environments,it is of great significance to plan feasible and highly safe driving paths for unmanned target vehicles.This paper focuses on the path planning and tracking control of unmanned target vehicles within accessible areas in off-road environments.To achieve this goal,based on the selected experimental area environmental model,suitable path planning algorithms are chosen for simulation verification.Additionally,a trajectory tracking controller is designed for tracking control validation.The main contents of this paper are arranged as follows:Part 1: Off-road Environment Modeling.Before conducting path planning,it is necessary to understand environmental information such as terrain and landforms.Through environmental modeling,this information is digitized and processed to create detailed maps,providing essential information for path planning.The selected experimental area in this paper is located in the northwest part of Liuhe District,Nanjing City.In order to gain a better understanding of the geographical conditions in this area,this paper utilizes BP neural network technology to classify the land cover types and extract and process the DEM elevation data of the experimental area.By constructing a map model using these parameters,the paper provides fundamental data support for subsequent path planning research.Part 2: Unmanned Target Vehicle Path Planning.Unmanned target vehicles navigate through complex off-road environments,requiring the assurance of feasible and safe driving paths.Therefore,the primary task of path planning is to determine a safe and feasible path from the starting point to the destination.Additionally,considering the time constraints of the unmanned target vehicle’s mission,the length of the path should also be taken into account during path planning.The goal is to minimize the length of the planned path while ensuring its feasibility and safety.For global path planning,an improved A* algorithm is employed in this paper to generate the desired path for the unmanned target vehicle.The algorithm takes into consideration terrain factors by modifying the cost function and using an adaptive weighting method to optimize the traditional A* algorithm,thereby speeding up the path search process.Simultaneously,to enhance path accuracy and driving efficiency,the Dynamic Window Approach(DWA)algorithm is utilized for local path planning.Real-time environmental information is used for path re-planning.Considering the terrain attributes,a reasonable speed plan is incorporated,resulting in an unmanned target vehicle trajectory that includes coordinates,heading,and velocity information.Part 3: Trajectory Tracking Controller Design.This article establishes a three-degree-of-freedom vehicle dynamics model and proposes a comprehensive control framework for unmanned target vehicles based on this model.By using a model predictive control algorithm-based lateral controller,lateral error control is achieved,while longitudinal speed control is achieved through a proportional-integral-derivative(PID)algorithm-based longitudinal controller.Finally,the effectiveness of the trajectory tracking controller is validated through straight-line and double-lane-change scenarios.Part 4: Simulation Validation of Off-Road Map Local Path Planning.Building upon the aforementioned foundation,this section combines the paths generated by the path planning algorithm in actual off-road environments with the designed trajectory tracking controller.The combined system is simulated and validated using the Matlab/Simulink and Carsim integrated simulation platform.The simulation results demonstrate that the comprehensive control trajectory tracker,in both lateral and longitudinal directions,can effectively control the vehicle to follow the predetermined trajectory in real-time,exhibiting excellent control performance.Furthermore,this validates the practical feasibility of the planned path and confirms its compliance with the safety requirements for unmanned target vehicle operations.In conclusion,this article has verified the rationality of the designed model and achieved the path planning and tracking control of unmanned target vehicles in off-road environments.This work provides a reference for the development and application of autonomous driving technology in special environments.