A UAV Path Planning Method Based on q(λ) Algorithm

Abstract

The invention provides a UAV mission planning method based on the Q(λ) algorithm, which includes an environment modeling step, a Markov decision process model initialization step, a Q(λ) algorithm iterative calculation step, and calculates the best value based on the state value function. In the optimal path step, first initialize the grid space according to the minimum track segment length of the UAV, map the grid space coordinates to waypoints, and represent circular and polygonal threat areas, and then establish a Markov decision model, including no Human-machine flight action space representation, design of state transition probability, construction of reward function, and then using the Q(λ) algorithm to perform iterative calculations based on the constructed model, and based on the final converged state value function, a path that can safely avoid To determine the optimal path for UAVs in threat areas, this invention combines the traditional Q-learning algorithm with utility tracking to improve the speed and accuracy of value function convergence, guide UAVs to avoid threat areas and perform autonomous path planning.

Description

technical field [0001] The invention relates to an unmanned aerial vehicle, specifically a path planning method for an unmanned aerial vehicle, belonging to the technical field of heuristic algorithms. Background technique [0002] UAV path planning is an important part of UAV mission planning and an important stage to realize autonomous mission execution of UAV. UAV path planning requires that in an environment with known, partially known or completely unknown information, it is planned to reach the target point from the starting point, which can bypass the threat area and obstacles, be safe, reliable and collision-free, and meet various requirements at the same time. A flight path with constraints. According to the acquisition of the battlefield environment information where the UAV is located, the path planning is divided into global path planning and local path planning. [0003] In practical applications, if the UAV can acquire global environmental knowledge, dynamic ...

AI Technical Summary

Problems solved by technology

[0004] The current local path planning technology has problems such as the algorithm is easy to fall into local minimum or local oscillation, the time cost of the algorithm is large, the computer information storage is large, and the rules are difficult to determine.

The behavior-based UAV path planning method is called the hotspot of current research. Its essence is to map the environmental state perceived by the sensor to the action of the actuator. In the behavior-based method, the design of the state feature vector and the supervised sample It is often very difficult to obtain in the actual complex environment

Images