针对当前路径规划算法在高维复杂环境下,存在障碍物信息处理不充分、搜索效率与路径质量难以平衡等问题,提出了一种基于自适应概率偏差的改进算法——APB-RRT*算法(Adaptive Probability Bias RRT*)。首先依据障碍物类型设定安全距离,...针对当前路径规划算法在高维复杂环境下,存在障碍物信息处理不充分、搜索效率与路径质量难以平衡等问题,提出了一种基于自适应概率偏差的改进算法——APB-RRT*算法(Adaptive Probability Bias RRT*)。首先依据障碍物类型设定安全距离,在保障安全的基础上将采样空间划分为m×m×m个子区域,引入参考路径邻近性评价函数,动态调整子区域采样概率,加快了算法的收敛速度,减少了冗余节点。同时,融合基于环境感知的自适应偏置策略和变步长调节机制,提高了路径搜索效率。最后,为遵循机械臂运动学特性,采用分段贪婪算法结合三次样条插值法对路径进行平滑约束,确保了路径的连续性和可执行性。仿真实验结果表明,该文算法规划出的路径最接近理想最优路径。展开更多
An improved RRT∗algorithm,referred to as the AGP-RRT∗algorithm,is proposed to address the problems of poor directionality,long generated paths,and slow convergence speed in multi-axis robotic arm path planning.First,a...An improved RRT∗algorithm,referred to as the AGP-RRT∗algorithm,is proposed to address the problems of poor directionality,long generated paths,and slow convergence speed in multi-axis robotic arm path planning.First,an adaptive biased probabilistic sampling strategy is adopted to dynamically adjust the target deviation threshold and optimize the selection of random sampling points and the direction of generating new nodes in order to reduce the search space and improve the search efficiency.Second,a gravitationally adjustable step size strategy is used to guide the search process and dynamically adjust the step-size to accelerate the search speed of the algorithm.Finally,the planning path is processed by pruning,removing redundant points and path smoothing fitting using cubic B-spline curves to improve the flexibility of the robotic arm.Through the six-axis robotic arm path planning simulation experiments on the MATLAB platform,the results show that the AGP-RRT∗algorithm reduces 87.34%in terms of the average running time and 40.39%in terms of the average path cost;Meanwhile,under two sets of complex environments A and B,the average running time of the AGP-RRT∗algorithm is shortened by 94.56%vs.95.37%,and the average path cost is reduced by 55.28%vs.47.82%,which proves the effectiveness of the AGP-RRT∗algorithm in improving the efficiency of multi-axis robotic arm path planning.展开更多
文摘针对当前路径规划算法在高维复杂环境下,存在障碍物信息处理不充分、搜索效率与路径质量难以平衡等问题,提出了一种基于自适应概率偏差的改进算法——APB-RRT*算法(Adaptive Probability Bias RRT*)。首先依据障碍物类型设定安全距离,在保障安全的基础上将采样空间划分为m×m×m个子区域,引入参考路径邻近性评价函数,动态调整子区域采样概率,加快了算法的收敛速度,减少了冗余节点。同时,融合基于环境感知的自适应偏置策略和变步长调节机制,提高了路径搜索效率。最后,为遵循机械臂运动学特性,采用分段贪婪算法结合三次样条插值法对路径进行平滑约束,确保了路径的连续性和可执行性。仿真实验结果表明,该文算法规划出的路径最接近理想最优路径。
基金supported by Foundation of key Laboratory of AI and Information Processing of Education Department of Guangxi(No.2022GXZDSY002)(Hechi University),Foundation of Guangxi Key Laboratory of Automobile Components and Vehicle Technology(Nos.2022GKLACVTKF04,2023GKLACVTZZ06)。
文摘An improved RRT∗algorithm,referred to as the AGP-RRT∗algorithm,is proposed to address the problems of poor directionality,long generated paths,and slow convergence speed in multi-axis robotic arm path planning.First,an adaptive biased probabilistic sampling strategy is adopted to dynamically adjust the target deviation threshold and optimize the selection of random sampling points and the direction of generating new nodes in order to reduce the search space and improve the search efficiency.Second,a gravitationally adjustable step size strategy is used to guide the search process and dynamically adjust the step-size to accelerate the search speed of the algorithm.Finally,the planning path is processed by pruning,removing redundant points and path smoothing fitting using cubic B-spline curves to improve the flexibility of the robotic arm.Through the six-axis robotic arm path planning simulation experiments on the MATLAB platform,the results show that the AGP-RRT∗algorithm reduces 87.34%in terms of the average running time and 40.39%in terms of the average path cost;Meanwhile,under two sets of complex environments A and B,the average running time of the AGP-RRT∗algorithm is shortened by 94.56%vs.95.37%,and the average path cost is reduced by 55.28%vs.47.82%,which proves the effectiveness of the AGP-RRT∗algorithm in improving the efficiency of multi-axis robotic arm path planning.
