Research On Collaborative Planning And Control Of Heterogeneous Multi-ground Unmanned Platforms

With the rapid development of industries such as smart cars,mobile robots,warehousing and logistics,cargo transportation,exploration and testing,more and more ground platforms are being used in various industries,such as smart disinfection mobile robots equipped with disinfection equipment used in epidemics protection,the coordinated transportation of multiple handling robots in the warehousing and logistics system,the coordinated transportation of parts by multiple vehicles in the factory,and the overall transportation of a large building in a certain formation for the preservation of ancient buildings,etc.,the above single platform may not be completed,or it will spend high costs,so many universities and enterprises have carried out a lot of research on multi-ground platforms,and their research directions include multi-platform task allocation,autonomous decision-making,trajectory planning,collaborative control,formation control,cluster transformation,collaborative obstacle avoidance,etc.aspect.This paper mainly conducts the following research work for the multi-ground platform collaborative system:(1)Combining the scenario of multi-vehicle collaboration,a three-layer system control framework is proposed,which are task distribution layer,trajectory planning layer and multivehicle control layer.The task distribution layer is responsible for the target position of each vehicle,so that the vehicles form a certain formation;the trajectory planning layer is responsible for vehicle path planning and speed planning,as well as obstacle avoidance strategy planning,etc.;the motion control layer is responsible for vehicle formation control strategies and trajectory tracking strategy.(2)Research the multi-vehicle task allocation algorithm,establish the vehicle model and communication model,and design and solve the optimal trajectory based on the Dubins curve.First,calculate the Dubins curve and the Dubins curve of any scheme according to the current position and speed information of the vehicle.The time for all vehicles to reach the target point,whichever is the longest.Use this time as the basis to predict the formation and target point of the vehicle after that time,and then find the optimal trajectory of the vehicle based on this target point,and complete the outermost Optimal trajectory;then speed planning is carried out according to the path,and the multi-objective genetic algorithm is applied to solve the speed sequence of multiple vehicles.The cost function is the distance and stability of the vehicles.Finally,the algorithm is simulated with five vehicles,and the vehicles change from triangles.It is a font to verify the effectiveness of the algorithm.(3)Research the vehicle motion planning algorithm,firstly analyze a variety of vehicle path planning methods,select the D*Lite algorithm for path planning,and improve the driving direction and optional grid range to increase,adjacent the cost of the grid takes into account the driving direction of the vehicle.If the driving direction is changed,the cost is low.On the contrary,the cost is large.Expanding the grid to eight directions will make the path smoother.Then a corresponding simulation was carried out for this algorithm to show its effectiveness and feasibility.(4)Research on multi-vehicle formation control algorithms,introduce nonlinear and linear model predictive control methods,and perform linear discretization of vehicle models and pilot follower models to construct a linear MPC framework;improve the cost function,and its objective function is mainly consider the vehicle trajectory error and its formation error;then the constraint conditions for collision avoidance between vehicles are studied,the VO obstacle avoidance method is improved,and its inequality constraint is obtained,combined with the previous framework to design an overall formation strategy;finally,the formation algorithm Perform simulation verification in matlab to prove the effectiveness and rationality of the algorithm.(5)Analyze the obstacle scene of multi-vehicle formation driving,and propose two formation obstacle avoidance schemes.Process analysis of the two schemes and the problems to be solved are carried out respectively.The first scheme includes three stages,namely transformation,Maintaining and restoring requires maximum distance analysis and formation transformation analysis.The second plan requires trajectory planning;combined with the previous analysis to conduct obstacle avoidance strategy research,the calculation method of formation error and path length two evaluation indicators is proposed.Trajectory planning was carried out for the two schemes;finally,obstacle avoidance strategy,trajectory planning and formation switching simulation were carried out to verify the effectiveness of obstacle avoidance strategy,trajectory planning strategy and the overall system framework.