Research On Motion Control And Motion Planning Of Quadrotor UAV

In recent years,with the miniaturization of electronic components and the continuous reduction of production costs,quadrotor unmanned aerial vehicles(UAVs)have been widely used in fields such as aerial photography,agricultural crop protection,and logistics delivery due to their excellent maneuverability and simple control requirements.With the continuous enrichment of the application fields of quadrotor,their working scenarios have gradually shifted from high altitude with fewer obstacles to complex near-ground environments,requiring UAVs to have robust autonomous navigation systems and strong perception and obstacle avoidance capabilities,which has become a research hotspot in the UAVs field,and there are still many challenging issues to be explored.On the one hand,as a statically unstable system,quadrotor have characteristics such as nonlinearity,under-actuation,strong coupling,and multivariable,and their small mass makes them susceptible to external factors such as gusts,making the design and implementation of high-quality motion control modules face great challenges.On the other hand,for the motion planning module of the UAV,its solution quality largely determines the quality of the UAV’s actual flight trajectory,but the computing resources that the UAV can carry are limited.When the UAV performs autonomous navigation flight in unknown complex environments,it is difficult to balance the optimality of the planned trajectory and the real-time planning efficiency.To address the above issues,this thesis mainly focuses on the motion planning and motion control problems of quadrotor in unknown environments,and conducts research from the aspects of the control model of the quadrotor,motion control based on nonlinear control algorithm,motion planning based on hybrid methods,and the software and hardware implementation of autonomous navigation systems.Considering the advantages and disadvantages of model-based and data-driven nonlinear control methods,as well as the relationship between real-time and solution quality of trajectory planning,under limited hardware facilities,this thesis designs a flight control method with certain antiinterference ability and tracking control accuracy and an efficient and high-quality trajectory planning solution method.Firstly,the nonlinear model of the quadrotor is studied in-depth,and based on the Newton-Euler equation,the nonlinear kinematic model and nonlinear dynamic model of the quadrotor are established,and the open-loop response characteristics of the system are studied as the basis for the design of the motion control module.Secondly,in response to the problem that the quadrotor has low trajectory tracking accuracy and unstable attitude control under external disturbances,a “positionattitude” cascaded control strategy is adopted,in which the position control loop uses a state space predictive control algorithm based on error to improve the trajectory tracking accuracy,and the attitude loop uses a active disturbance rejection control algorithm to resist unknown external disturbances.For the local real-time planning problem of quadrotor in unknown environments,a hybrid motion planning framework is adopted,which uses a front-end motion planning algorithm based on Hybrid A* to improve trajectory search efficiency,and a back-end trajectory optimization algorithm based on B-spline curves to improve the quality of the initial trajectory.Finally,flight test evaluations are conducted on the motion control algorithm and motion planning algorithm designed in this thesis,including attitude control test and outdoor autonomous navigation flight test,and based on the test data,the effectiveness of the attitude control and the autonomous navigation trajectory are evaluated.