A new dynamic path planning method in high dimensional workspace, radial based probabilistic roadmap motion (RBPRM) planning method, is presented. Different from general probabilistic roadmap motion planning methods, ...A new dynamic path planning method in high dimensional workspace, radial based probabilistic roadmap motion (RBPRM) planning method, is presented. Different from general probabilistic roadmap motion planning methods, it uses straight lines as long as possible to construct a path graph, so the final path obtained from the graph is relatively shorter and straighter. Experimental results show the efficiency of the algorithm in finding shorter paths in sparse environment.展开更多
针对矿难发生后井下环境的不确定性,提出一种以矿难前的GIS(Geographic information system)地图为基础建立环境栅格模型并结合改进遗传算法的矿难搜索机器人全局路径规划方法。效仿蚁群算法中的信息素提出基于位置信息负反馈的方法,并...针对矿难发生后井下环境的不确定性,提出一种以矿难前的GIS(Geographic information system)地图为基础建立环境栅格模型并结合改进遗传算法的矿难搜索机器人全局路径规划方法。效仿蚁群算法中的信息素提出基于位置信息负反馈的方法,并结合优先权分组的思想,提出一种新的有效的种群初始化方法,同时将该种群初始化方法应用到变异算子中,且依据最优解的变化情况自适应地调整交叉和变异的概率。与此同时,针对环境信息的不同变化情况,结合全局路径规划结果对机器人进行局部避障方法的研究。最后,通过仿真实验证明本方法能够快速有效地在已知环境中得到机器人的最优路径,并且能够在局部变化的环境中实现实时避障。展开更多
文摘A new dynamic path planning method in high dimensional workspace, radial based probabilistic roadmap motion (RBPRM) planning method, is presented. Different from general probabilistic roadmap motion planning methods, it uses straight lines as long as possible to construct a path graph, so the final path obtained from the graph is relatively shorter and straighter. Experimental results show the efficiency of the algorithm in finding shorter paths in sparse environment.
文摘针对矿难发生后井下环境的不确定性,提出一种以矿难前的GIS(Geographic information system)地图为基础建立环境栅格模型并结合改进遗传算法的矿难搜索机器人全局路径规划方法。效仿蚁群算法中的信息素提出基于位置信息负反馈的方法,并结合优先权分组的思想,提出一种新的有效的种群初始化方法,同时将该种群初始化方法应用到变异算子中,且依据最优解的变化情况自适应地调整交叉和变异的概率。与此同时,针对环境信息的不同变化情况,结合全局路径规划结果对机器人进行局部避障方法的研究。最后,通过仿真实验证明本方法能够快速有效地在已知环境中得到机器人的最优路径,并且能够在局部变化的环境中实现实时避障。
