Research On Obstacle Avoidance Planning And Trajectory Tracking For Deicing Robot On Transmission Lines

In recent years, the bad weather conditions in winter will result in snow and icingon the transmission lines, which causes some disasters like line trip, pole collapse,insulator flashover, communication interruption and so on. It not only gives rise tohuge nationally economic losses, but also brings inconvenience to people’s life.Therefore, it has an urgent demand and great significance to research and develop thedeicing robot on transmission lines. Considering the special working environment ofdeicing robot, obstacle avoidance planning and trajectory tracking are studied in thisdissertation. The main studies undertaken in this thesis are listed as follow:Thesis first introduces the research background and significance of deicing robotsystematically. Then the research progresses of deicing robot both at home and abroadare reviewed, and difficulties and key technologies of deicing robot are analyzed indetail as well. In the second section, a self-developed deicing robot with three arms isintroduced, including mechanical structure, control system structure and its hardwareselection and software design. Then, the main research objects, a theoreticalkinematic robotic arms model and another dynamic one, are established by MATLAB.When robotic arms move, its action performance is affected by weight of roboticarms besides a certain workload, which performs nonlinear characteristic and brings alot of problems for control. While traditional fixed value of gravity compensation andapproximation of gravity compensation have some disadvantages, such as inaccuratecompensation and poor real-time performance. Therefore both of these compensationmethods can not meet the requirement of robotic arms control. At this point, anaccurate online gravity compensation strategy based on the space geometry and theoryof mechanics is proposed. By combining it with PD control law, an online precisegravity compensation of PD control law for deicing robot manipulators is presented,whose stability is proved by Lyapunov theorem. Finally simulation verified theeffectiveness and superiority of proposed control law.On the basis of stability control and good tracking ability of deicing robotrobotic arms, an obstacle avoidance planning strategy based on fuzzy logic ispresented, in order to meets the needs of overcoming barriers when deicing robotworking on transmission lines. In this strategy, surroundings information is collectedby lidar. And error information of robotic arms ending coordinates and obstacle judgement are used as fuzzy planning inputs. In this way, work state can be dividedinto safe mode and dangerous mode, and different joint angles are planned in differentmode. In the end, simulation results show that robotic arms move effectively both intwo modes, especially shows a good performance on obstacle negotiation indangerous mode.In addition, considering the redundancy problem of fuzzy planning trajectory,particle swarm optimization is introduced to optimize the fuzzy planner parameters,then an obstacle avoidance planning strategy based on fuzzy logic and particle swarmoptimization is proposed. In this strategy, particle swarm optimization is used foroptimizing joint angles in real time by designing a fitness function of shortest pathand target attraction. Through the comparison of the simulation experiments, theeffectiveness of the approach is verified to be able to reduce the redundancy ofplanning trajectory, and good for deicing robot working online for a long time.