Research On Dynamic Path Planning Of Multi-rotor UAV In Indoor Environment

Multi-rotor UAV have attracted extensive attention from researchers in the field of artificial intelligence as mobile robots with strong autonomous behavior.At present,the application environment of UAV path planning at home and abroad is generally outdoor open area,while the indoor obstacles are complex in shape,changeable in distribution mode and more diversified in movement mode.Therefore,the path planning algorithm of indoor drones is studied to realize the real-time acquisition of pose information in the complex and variable environment of the UAV and to fly along the expected value of the trajectory,which will lay a solid foundation for the intelligent autonomous navigation technology of the UAV.Important theoretical research significance and application reference value.Based on the traditional grid method,this paper proposes a three-dimensional environment modeling method with high dimensionality reduction,which reduces the number of grids in the traditional grid method,improves the accuracy of environmental modeling,and reduces the computational complexity of the planning algorithm.It satisfies the efficiency requirements of the planning algorithm and makes up for the shortcomings of the traditional grid method.When using the grid method for environmental modeling,the number of grids will be greatly increased compared to the two-dimensional environment,and the efficiency of the planning algorithm will be greatly reduced.Therefore,solving the problem requires a condition that can ensure the efficiency of the planning.The planning algorithm that satisfies the environmental accuracy requirements.In this paper,a Non-memory-regression A*algorithm is proposed to perform dynamic path search based on global path planning using A*algorithm to solve the dynamic global path planning problem in indoor three-dimensional space environment.In the global path planning process of the multi-rotor UAV in the indoor environment,although a large amount of global environmental information can be obtained by the current technical means,when the path planning is performed in the indoor three-dimensional environment,the indoor space is relatively small,and the obstacle avoidance is relatively small.The degree is small,the obstacles are irregular and complex,so a planning algorithm that can ensure the planning efficiency and meet the environmental accuracy requirements is needed.In this paper,the A*algorithm is used as the global path planning algorithm to overcome the complex variability of indoor three-dimensional space environment information.The effectiveness and real-time performance of A*algorithm in global path planning are verified by simulation experiments.In the actual flight process.multi-rotor UAV must conduct autonomous flight control in an information-diversified environment,and also deal with complex and variable environmental conditions.This type of problem is difficult to solve by using the global planning method.In the face of complex environments,a more real-time path planning method should be utilized.Since the Non-memory A*algorithm has good performance in dynamic path planning,dynamic programming is easy to fall into local optimum when the environment is complex and variable.The Non-memory-regression A*algorithm proposed in this paper overcomes the locality and blindness of the planning algorithm in the process of dynamic path planning.Finally,the simulation results verify the correctness and efficiency of the dynamic path planning without Non-memory-regression A*algorithm.In this paper,the UAV indoor dynamic path planning flight test platform is designed and built,the selection scheme and hardware construction process of multi-rotor UAV are determined,and the Qt-based ground station system is developed.The visual positioning platform and ground are designed and built.The communication interactive link between the station system and the multi-rotor UAV validates the validity and real-time performance of the Non-memory-regression A*algorithm in the multi-rotor UAV in the indoor environment under the indoor experimental platform.