The Route Planning And Robust Control For Rotor UAVs

A autonomous flight system is designed for a RUAV(Rotorcraft Unmanned Aerial Vehicle)operating in complex environment,making the RUAV successfully avoid static and dynamic sudden obstacles and reach the destination at a lower cost in complex environment.And can guarantee the stable flight of the RUAV.In order to achieve this function,the paper mainly studies the two aspects of path planning and robust intelligent controller.Among them,the design of the combination of static and dynamic path planning algorithm provides the real-time desired position for the RUAV,and at the same time designing the high-performance intelligent controller to control the RUAV to achieve accurate and fast tracking of the fixed point,so that the RUAV can stable fly along the better path from the starting point to safely reach the target point,and is free from external interference to achieve stable flight and task operating,which provides some help for the real “unintelligent” and “unmanned” operation of the RUAV.The main work of the paper is as follows:1.The static path planning based on AC-PSO fusion algorithm is proposed.Aiming at the shortcomings of the classical ant colony algorithm in path length and smoothness,the node movement rules of ant colony algorithm are modified,multiple heuristic information is constructed,and the particle swarm optimization algorithm is combined to compensate for the global scope of the search range.The AC-PSO(Ant Colony-Particle Swarm Optimization)fusion algorithm greatly improves the perform ance of path length and smoothness,and can plan a low-cost,smooth and safety path in a known static environment,and as a reference path for dynamic path planning later.2.The dynamic path planning based on neural dynamics model algorithm is proposed.In the classical neural dynamic model update equation,the target direct influence term is added,the cubic function is proposed as the connection weight function,and the track planning process structure is changed.Following the idea of combining static and dynamic track planning,the RUAV will fly along the reference path by default.For the sudden obstacles or changing target points that affect the advancement,the neural dynamic model algorithm is activated for dynamical plan and online adjustment to achieve dynamic obstacle avoidance and target tracking.3.The position-attitude flight controller for RUAV is designed based on adaptive RBFNN\ADRC algorithm.Firstly,the adaptive RBFNN algorithm is used to design the interference observer to realize the online estimation of unk nown interference.Then based on the self-disturbance control theory,combined with tracking differentiator,adaptive RBFNN noise measurement and nonlinear state error feedback,the adaptive RBFNN\ADRC(Active Disturbance Rejection control algorithm based on adaptive Radical Basis Function Neural Network)is designed.The algorithm is used to construct a position-attitude double closed-loop controller to realize stable flight of the RUAV.The flight controller is suitable for RUAV operating in complex environments,and can effectively suppress internal and external disturbances to make it work stably.Finally,combining path planning module and the intelligent flight control module to build RUAV autonomous flight system.First,according to the combinative ideals of static and dynamic path planning.Using static reference path and dynamic planning online adjustment to implement dynamic obstacle avoidance and target tracking,while providing real-time desired position for the RUAV,and controlling the RUAV to achieve accurate and fast tracking of the fixed point through the flight controller based on adaptive RBFNN\ADRC,so that the RUAV can fly following the short-length and well-smoothness path from starting point to the target point safely and ensuring that the RUAV can fly stably and perform tasks under external interference.