Abstract
Obstacle avoidance path planning capability, as one of the key capabilities of UAV (Unmanned Aerial Vehicle) to achieve safe autonomous flight, has always been a hot research topic in UAV research filed. As a commonly used obstacle avoidance path planning algorithm, RRT (Rapid-exploration Random Tree) algorithm can carry out obstacle avoidance path planning in real time and online. In addition, it can obtain the asymptotically optimal obstacle avoidance path on the premise of ensuring the completeness of probability. However, it has some problems, such as high randomness, slow convergence speed, long transit time, and curved flight trajectory, so that it cannot meet the flight conditions of the actual UAV. To solve these problems, the paper proposes an improved RRT algorithm. In the process of extending the random tree, ACO (ant colony optimization) is introduced to make the planning path asymptotically optimal. The optimized algorithm can set pheromones on the path obtained by RRT and select the next extension point according to the pheromone concentration. And then through a certain number of iterations, it converges to an ideal path scheme. In addition, this paper also uses MATLAB to verify the effectiveness and superiority of the algorithm: Although RRT is easy to fall into local optimization, since the optimization method in this paper can almost certainly converge to the optimal solution, when it is necessary to preplan the path before UAV takeoff, it can be used.
Highlights
Planning on autonomous cruise of UAV (Unmanned Aerial Vehicle) with a collision-free track, which meets the high requirements about continuity constraints and output stability, is the research hotspot of multiagents’ intelligence
When the first-generation m ants find m paths according to RRT, the pheromone concentration of all grid points on the map can be updated according to the above method
Because the result of RRT algorithm is basically not affected by the complexity of the environment, this paper first thought of using it
Summary
Planning on autonomous cruise of UAV (Unmanned Aerial Vehicle) with a collision-free track, which meets the high requirements about continuity constraints and output stability, is the research hotspot of multiagents’ intelligence. Taheri et al [18] proposed a path planning method that can effectively reduce the running time and computational complexity, which is called fuzzy greedy fast exploration random tree (FG-RRT). Wang et al [19] modelled the tree selection process as a multiarm bandit problem and used reinforcement learning algorithm to learn action values It can enhance the local space exploration ability of each tree, and ensure the efficiency of global path planning. Considering that ACO can always find the best path after enough iterations, it can be used to optimize the path obtained by RRT and get better results. ACO is introduced and an RRT path optimization method for ACO is proposed in this paper
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
