Research On Path Planning And Collision Avoidance Control Of Unmanned Surface Vehicle

Unmanned Surface Vehicle(USV)has attracted extensive attention in military and civil fields because of its small volume and fast speed.In the complex and changeable marine environment,it is an important content of USV technology research to ensure that USV can safely avoid obstacles and sail to the end of the mission.The research on path planning and collision avoidance control technology for USV is the key technology to realize the autonomous navigation of USV.In this dissertation,the path planning and collision avoidance control technology are studied mainly for obstacle constraints,maneuvering characteristics constraints such as dynamic vessel model and collision avoidance rules constraints.In this dissertation,the following aspects are mainly studied:1.The maneuvering motion model of USV is established.According to the propulsion characteristics of single stern engine,the propeller speed and propulsion angle are used as control inputs.Taking the Lanxin USV as the experimental platform,the real experimental data are obtained through zig-zag test,turning test and straight acceleration test.The parameters of the USV model are identified by multiple-input multiple-output recursive least square method,and the USV motion model with maneuvering characteristics constraints is established,which is suitable for the research of path planning and collision avoidance control.2.A dynamic path planning algorithm based on state constraints is proposed.The algorithm can obtain the optimal planned path between the starting point and end point,and establish the corresponding ship domain model and state constraint model to constrain the planned path according to the state information of moving obstacles and own vessel,so that the planned path can realize autonomous planning according to the constraints of obstacles and own vessel state.At the same time,the planned path can be adjusted according to the requirements of safety margin and path length.Under the condition of meeting the motion state constraints of USV,the algorithm can independently plan a continuous,smooth and safe path,to ensure the feasibility of USV tracking the planning path.3.A real-time collision avoidance control algorithm based on dynamic constraints is proposed.The algorithm establishes the candidate control behavior space of propeller speed and propulsion angle,and obtains the dynamic feasible prediction trajectories according to the maneuvering motion model of USV.The evaluation function of collision avoidance control is constructed.The evaluation function comprehensively considers the factors such as collision risk assessment conditions,moving obstacle constraints and sailing stability.Through the evaluation function,the optimal control quantities are obtained in real time and the planned trajectory is generated.On the premise of fully considering the maneuvering characteristics of USV,the optimal maneuvering control quantities and the generated planning trajectory obtained can meet the constraints of moving obstacles,dynamics,actuators and collision avoidance rules at the same time,which overcomes the estimation problem of velocity window parameters for dynamic window collision avoidance algorithm.4.Aiming at the problem that the calculation speed of dynamic path planning algorithm is slow and it is difficult to make rapid update response to sudden changes in the marine environment,and the real-time collision avoidance control algorithm is easy to fall into local optimization only considering the local environmental constraint information,an path planning and collision avoidance control method with interactive operation of dynamic path planning algorithm and real-time collision avoidance control algorithm is proposed.The proposed path planning and collision avoidance control method relies on the dynamic path planning algorithm to ensure the global optimization and consistency of the planned trajectory,and relies on the real-time collision avoidance control algorithm to quickly respond to the changes of local environmental constraint information,so that they can make up for their shortcomings and update each other.Thus,the dynamic feasible planning trajectory satisfying various constraints between the starting point and the end point in the complex environment is generated,and the optimal control quantities are obtained.