Aircraft Taxiing Path Optimization Algorithm Based On Conflict Hotspots

In recent years,with the rapid development of China ’ s civil aviation industry,aircraft in the process of multi runway airport scene taxiing frequently conflict collisions and runway intrusions and other events,resulting in an increase in the rate of flight delays,seriously affecting the efficiency and safety of the airport.How to reasonably plan the aircraft taxiing path and efficiently complete the conflict resolution strategy is one of the most concerned issues at present in domestic airports.Based on the above problems,this paper proposes an aircraft taxiing path planning algorithm based on conflict hotspot to improve aircraft taxiing efficiency in complex scene environment.The main work and innovation of this paper are as follows: Obtain real data and generate simulation flight plan.Combined with Monte Carlo simulation,the prediction and identification of conflict hotspots are completed.The advantages between the path planning algorithm based on improved Q-learning algorithm and the path planning algorithm based on improved A*algorithm are analyzed.On this basis,the conflict hotspots are used to make reasonable choices for the scene path planning strategy.1、This paper analyzes the interaction between the various systems in the operation area of multi-runway airports,completes the modeling of the operation area of multi-runway airports through the node-path method and constructs the corresponding directed graph.The air traffic control simulator parses and imports the real data of the airport scene and constructs the simulation flight plan.At the same time,the Monte Carlo simulation strategy is used to predict and identify the conflict hotspots.The potential conflict hotspots that are not identified are detected,and the conflict hotspots are classified.2、In order to improve the efficiency of aircraft taxiing on the scene,this paper proposes two static path planning strategies,namely,static taxiing path planning strategy based on improved Q-learning and static taxiing path planning strategy based on improved A* algorithm.The path planning algorithm based on improved Q-learning improves the efficiency of aircraft taxiing on the scene by introducing the dynamic search factor ε.The path planning algorithm based on improved A* avoids the occurrence of head-to-head conflict from the root.At the same time,the conflict area will be concentrated in some hot spots to reduce the control load,and it is also convenient for controllers to master the taxiing path.Most importantly,this sliding optimization method can reduce the frequency of conflict hotspots and improve the security of scene operation.On the basis of the above two static path planning algorithms,a dynamic path planning algorithm based on conflict hotspot level assignment is proposed.When the sum of airport conflict levels is higher than the minimum threshold,the path planning algorithm based on improved Q-learning is adopted.Otherwise,the path planning algorithm based on improved A* algorithm is adopted.When aircraft conflicts occur during taxiing,the Q-learning algorithm is used to achieve conflict resolution.3、This paper verifies and analyzes the effectiveness of path planning algorithm based on improved Q-learning.At the same time,the advantages and disadvantages of the taxiing path planning algorithm based on improved Q-learning and the taxiing path planning algorithm based on improved A* algorithm are compared under different conflict levels.Through the dynamic selection of the scene path planning algorithm in the unit time period,the effectiveness of the path planning algorithm based on conflict hotspots is verified,which improves the taxiing efficiency of the airport scene and provides an auxiliary decision function for the regulators in the field control.In summary,the research work in this paper combines the scene modeling,data processing and reinforcement learning Q-learning,and explores new research strategies to improve the efficiency of airport scene taxiing,which provides an effective solution to improve the safety of the scene.At the same time,the method proposed in this paper is verified by experiments.