Research And Implementation Of Fast Motion Planning For Autonomous Navigation Robot

As China’s population ages and the birth rate decreases,there will be more and more elderly people and fewer young adults in the future,and social labor costs will increase greatly,which provides an extremely broad market for server robots.And autonomous navigation technology,as the core technology of server robots,will receive more and more attention.In this paper,we focus on the motion planning algorithm of mobile robots in indoor scenes,and design the path planning algorithm incorporating JPS and improved A~* algorithm and the minimum-jerk trajectory planning algorithm based on Bezier curve improvement,respectively.In this paper,an improved strategy of fusing the JPS jump point algorithm and the A~*algorithm is proposed to address the problems of traversing many redundant nodes leading to high memory consumption and slow computation speed of the pathfinding algorithm when the traditional A~* algorithm is used for path planning of raster maps with large factory scenes.The superiority of the improved algorithm is verified by analyzing and comparing the path planning of raster maps of different sizes.Simulation results show that the proposed algorithm not only speeds up the A~* algorithm to a great extent,but also consumes significantly less memory for the system runtime.In this paper,the trajectory planning is based on the minimum-jerk algorithm,which generates trajectories for localized paths based on the path points generated by front-end path planning.The minimum-jerk algorithm can generate a relatively smooth trajectory,but due to the problem of multi-segment trajectory time allocation,it is easy to cause velocity or acceleration overshoot,which is manifested in the trajectory as a significant deviation from the path point or even cause the phenomenon of knotting.To address this problem,firstly,in order to restrict the shape of trajectory generation,the space corridor is added to the quadratic programming problem as a constraint,then the solved trajectory is naturally within the space corridor.However,this also brings the problems of large computational effort and inconvenient to impose global area safety and dynamically feasible constraints.Finally,Bezier curve is used instead of the quintic polynomial in the minimum-jerk trajectory planning algorithm,and the above problems can be completely avoided and the computational effort can be reduced by using the characteristics of Bezier curve to improve the speed of trajectory planning.In this paper,a lidar-based indoor autonomous navigation robot platform is built and extensive experiments are conducted in an indoor environment.The experimental results show that the proposed path planning incorporating JPS and improved A~* algorithm and Bezier curve improved minimum-jerk trajectory planning algorithm can plan smooth and collision-free paths.The designed motion planning algorithm achieves better results in both static and complex dynamic environments.