Research On Path Planning And Autonomous Navigation Technology Based On FROPI

With the continuous expansion of China’s power grid scale and the gradual advancement of the power system reform,the traditional way of manual line inspection is increasingly unable to meet the needs of daily inspection of power lines.As a safe and efficient new technology,UAV Patrol has been widely used in power patrol,river patrol,high-speed patrol,meteorological monitoring and other fields in recent years.At present,UAV inspection is mainly completed by remote control of staff.If the autonomous flight of the UAV can be achieved,the inspection efficiency will be further improved.Path planning is the precondition of autonomous flight of the UAV,and autonomous navigation is the important guarantee of path planning.Therefore,the research of UAV path planning and autonomous navigation technology,and applying them to the daily inspection of power lines,has important significance and practical value for the detection of line faults,the improvement of inspection efficiency,and the stability of power grid system.Starting from the research status of UAV route inspection at home and abroad,this paper analyzes the key issues of UAV route inspection,determines the inspection constraints,and builds a safety inspection area model to ensure the safety of UAV and route in the process of inspection.On this basis,according to the different technical content,this paper discusses the route inspection of UAV in detail from two aspects of path planning and autonomous navigation.The main contents of this paper are as follows:First of all,the principle of RRT algorithm is introduced in detail.On this basis,several typical improved RRT algorithms are studied in depth.Through simulation experiments,the advantages and disadvantages of RRT algorithm are evaluated from three aspects: search efficiency,convergence speed and path quality.The experimental results show that RRT-Connect algorithm is more in line with the needs of UAV line inspection path planning.Secondly,according to the characteristics of narrow channel in the UAV safety inspection area model,a bridge detection algorithm for narrow channel sampling is selected.Aiming at the defect of bridge detection algorithm,an improved bridge detection algorithm is proposed.The original global sampling space is replaced by the set of obstacle edge nodes,which reduces the sampling range and improves the sampling efficiency;the bridge and its two endpoints are constructed by a simple local geometry testing method,which avoids complex Gauss calculation and effectively reduces the complexity of the algorithm;according to the distribution characteristics of narrow channel sample points,the Connect algorithm is adjusted to speed up the expansion of the sample points based on the distribution characteristics of the narrow channel sample points.Through simulation experiment,the narrow channel sampling performance of RRT-Connect algorithm alone and the improved bridge detection algorithm before and after is compared to verify its effectiveness.Finally,the basic principle of Beidou/INS integrated navigation system is briefly explained,and the mathematical model of the system under the loose combination mode is derived.Aiming at the problem that the traditional EKF algorithm is too simple to deal with the noise characteristics,which may lead to the increase of filtering error,an adaptive extended Kalman filtering algorithm is adopted.Through the residual of the filter system state,the system noise variance and measurement noise variance are adjusted in real time,and then the filter gain is changed to achieve better carrier tracking.The flight path of a UAV is simulated by MATLAB software.The experimental results show that the adopted algorithm is better than the traditional EKF algorithm in filtering accuracy.