Optimal Path Planning Combined With Obstacle Avoidance And Application Research On Delivery Trolley

With the rapid development of artificial intelligence in the past decade,the unmanned technology as an iconic achievement in the development of artificial intelligence is receiving more and more attention from scholars in all walks of life.Unmanned vehicles are a highly automated product that involves a lot of knowledge and it is a typical project of interdisciplinary subjects.As one of the unmanned technology landing industries,the unmanned delivery car plays a key role in improving the automation of logistics.Along with the high-speed development of the logistics industry,the factory logistics has become more and more multifarious,and the social demand and industrial demand for unmanned delivery cars are unprecedented.Path planning and obstacle avoidance technology are an indispensable part in the unmanned technology.In this paper,a series of related research algorithms are proposed for the path planning combined with obstacle avoidance control and the path optimization based on the shortest path-smoothing process,in order to increase the application of unmanned delivery vehicles in the logistics automation industry.Specifically,this dissertation conducts research from the following aspects:Firstly,by contrasting and analyzing several methods of environmental modeling and path planning,this paper uses the link graph method for environment modeling,while the path planning aspect selects Dijkstra algorithm and ant colony algorithm.After the environment model is built,the preliminary path planning is performed by Dijkstra algorithm to find a suboptimal path.Secondly,the ant colony algorithm is used to optimize the suboptimal path,and the optimized shortest path is obtained.At the same time,the related mathematical parameters and models of the ant colony algorithm are introduced in detail in the paper,and the relevant data obtained by the experiment are used to analyze the influence of these parameters on the performance of the algorithm.Lastly,a set of optimal parameters obtained in the experiment are applied to the path optimization.Next is about the path smoothing.Since the path optimized by the ant colony algorithm is segmented,the path needs to be smoothed.In this paper,by analyzing theBezier curve and the B-spline curve,it is found that after the Bezier curve,the smoothed line is prone to large deformation,and the smoothing effect of the B-spline curve is better.Therefore,this paper adopts B spline curve to smooth path planning.Finally,on the basis of existing path planning and optimization processing,combined with the specific situation of the delivery car model,the research on the path planning application of the delivery car is carried out.Based on the physical collision model of the delivery trolley,the obstacles in the environment are swelled,and the avoidance of the interference of the environmental obstacles by the actual model of the trolley is realized.Besides,based on the car movement model constraints,the model of the car based on the Mecanum wheel is analyzed to verify the feasibility of the planning algorithm and motion constraints.According to the above analysis results and proposed algorithm,the actual navigation environment is modeled and path planning is verified at the end of this chapter.The experimental results show that the proposed system method meets the obstacle avoidance requirements of the actual navigation environment and achieves effective optimization.