论文部分内容阅读
针对大部分反应式避障算法未考虑机器人尺寸约束导致机器人无法通过障碍物间可行通道的问题,提出了一种复杂环境下考虑机器人尺寸约束的动态窗避障方法。利用激光雷达提供的精确测距信息,通过探索机器人运动前方的可通行区域,避免陷入“U”型等局部最小区域;以障碍物间可行通道的宽度与机器人自身尺寸间的关系作为新的引导,构造目标函数,提高机器人对穿越路径上有限宽度可行通道的决策能力。仿真实验验证了该方法能避免局部最小问题、利于全局路径优化,提高了机器人在可行狭窄通道间的穿越避障能力,表明该方法具有有效性。
In order to solve the problem that most of the reactive obstacle avoidance algorithms do not consider the constraints of robot size and the robots can not pass through the obstacles between obstacles, a dynamic window obstacle avoidance method considering robot size constraints under complicated environment is proposed. Using the accurate distance measurement information provided by laser radar, we can avoid getting into the local minimum area such as “U ” type by exploring the accessible area in front of the robot motion. Taking the relationship between the feasible channel width of obstacle and the robot’s own size as a new The objective function is constructed to improve the decision-making ability of the robot to the feasible channel with limited width on the traversing path. Simulation results show that this method can avoid the local minimum problem, which is benefit to the global path optimization and enhances the obstacle avoidance ability of the robot crossing the feasible narrow channel. It shows that this method is effective.