Research On Three-dimensional Airline Planning Method In Autonomous Flight Drone Room

With the advancement of key technologies such as energy and power technology,aircraft platform technology,intelligent control technology,information transmission technology,drone in the function,application field,has gradually expanded from outdoor research and application to indoors,the autonomous flight and application research in the drone indoors have entered the rapid development stage and is gradually become a hot problem for current and future research.Compared with most drones,the advantages of four-rotorless drones are simple,smaller,easy to operate,and can replace humans to complete a variety of complex tasks such as risk,post-disaster rescue,indoor search,and other complex tasks.In actual production applications,the four-dimensional airplancular planning is especially important.It is directly related to the obstacle to the surrounding environment.This paper surrounds the autonomous flight in the four-rotorless chain,and proposes a higher efficiency of the track planning algorithm,and specific research on the three-dimensional airline planning of four-rotorless drone indoors.The main research work of the full text is as follows:(1)The grid method is used to establish an indoor three-dimensional model,and the problem of drone search direction is analyzed.Combined with the structure and flight characteristics of the quadrocriopersy drone,the flying conditions of the drone in the indoor environment were analyzed,the algorithm used in the track planning problems,and the advantages and disadvantages of various airline planning algorithms were analyzed and compared,select an ant colony algorithm and Quick-RRT*algorithm as an algorithm research object.(2)For the complex characteristics of indoor space,there is an initial blind search for the ant colony algorithm,the convergence speed is slow,and it is easy to fall into a local optimal problem.It is proposed to improve the ant colony algorithm,and improve the setting method of the initial pheromone,and enhance the ant colony that search direction;design inspiration probability,and improve status transfer rules,effectively improve the visibility accuracy of ant colony;optimize the update of pheromones,increase the pheromone to the dynamic adjustment strategy,accelerate the convergence speed of the algorithm,effectively avoid it follow the topically optimal.And through experiments,the algorithm verification is verified,and the results show that the optimized ant colony algorithm decreases by 38.63%compared with the traditional ant colony algorithm in terms of the average path length,the average use of the algorithm is 3.80%.Ant Colony Algorithm.In order to further prove the reliability of the ant colony algorithm in this article,compare it with the other two improved algorithms,and verify that the improvement algorithm of this article has good stability.(3)The problem of slow convergence at the Quick-RRT*algorithm is integrated with the improvement of gray wolf algorithm.Join the scold field function in the artificial potential field method to keep a certain safe distance between the planned trails and obstacles,ensure the safety of drone flight,and make smooth optimization of the trails obtained to make the obtained,the trails meet the actual flight requirements of the drone.Aiming at the sudden obstacles of drones on the flight trail,the local aircraft track re-planning strategy is used to carry out local re-planning treatment.The experimental results show that compared with the improvement of the gray wolf algorithm and the Quick-RRT*algorithm,the Adapted-RRT*algorithm convergence is accelerated,which enhances the ability of drones to find better trains and the ability to respond to emergencies,and satisfy satisfaction.Real-time requirements for flights planning.The experimental comparison results are:Adapted-RRT*algorithm decreases by 37.72%and 21.53%compared with the previous two algorithms,respectively,and the time is reduced by 40.21%,78.21%,respectively.