Research On Real-time Strategy Games Based On Flow Field Pathfinding Algorithm

Real-time strategy games are a type of games for resource collection,construction,and real-time combat.They are a very important type of video games.The characteristics of real-time,interactive,and playable has been required by real-time strategy games which determine that pathfinding,one of the important components of real-time strategy games,should satisfy the multi-agents' ability to search for paths and achieve rapid movement at the same time,and realistically and naturally pathfinding of simulation crowd in the game world.Although the traditional pathfinding algorithm realizes the shortest path,the number and speed of the pathfinding crowd cannot fulfil the player's requirements,which affects the player's gaming experience.In recent years,some scholars have applied artificial potential field method and intelligent bionic algorithm to game pathfinding,but they still cannot fulfil the needs of multiple agents in the real-time strategy game to reach the destination at the same time,and they still use an optimized single path.The speed of the path planning improved by the individual times,not overall to improve path planning speed of crowd.A large number of agents will experience fast,slow,arched congestion in the existing path planning methods during the movement process.Therefore,this paper focuses on the above problems to study the multi-agents intelligent pathfinding algorithm and simulated crowd model in real-time strategy games.The main work and innovations of this paper are as follows:1.Currently,the main pathfinding algorithm is to plan a path from the starting point to the end.It cannot solve the problem of multi-agents seeking paths simultaneously in a real-time strategy game and finding multiple paths concurrently.Aiming at this problem,this paper uses intelligent flow-field pathfinding algorithm to solve the problem of path planning,and proposes an improved flow-field pathfinding algorithm based on combination.The algorithm improves the algorithm from three aspects: data storage,complete generation value calculation and flow field generation.Based on the realization of concurrent pathfinding,it further reduces the pathfinding time and improves the intelligence level of the game.2.Although the flow-field pathfinding algorithm solves the problem of multi-agent concurrent pathfinding,the phenomenon of fast,slow,arched congestion in large-scale group simulation will still occur.Aiming at the problem of multi-agent path-finding collision blocking in game scenes,this paper combines a high-density simulated crowd model to improve the flow-field path-finding algorithm,so that the multi-agents in real-time strategy games can find paths in real time and simulate real-world movement.The model determines the next position of the agent from the psychological and geometric rules,and comprehensively analyzes the attractiveness of the target position,the repulsive force of obstacles such as walls,the forces exerted by other mobile agents,and the combined force of the previous forces to obtain intelligence.The direction and speed of the body's movement determine the next position and the algorithm adjust the parameters to the appropriate weight according to the rules,and finally determine the movement of the entire group.3.The algorithm and model are applied in the actual scene simulation,and the scenario of real-time pathfinding and movement of multiple agents is realized and the algorithm was compared with others.The algorithm achieved by the platform of the Visual Studio2017 and Open GL,and compared with the traditional pathfinding algorithm,heuristic algorithm and new pathfinding algorithm.The experimental results prove that the improved flow field pathfinding algorithm combined with high-density crowd simulation model is suitable for the scenario where multiple agents move instantly.