基于蜂窝地图的全方位步进式机器人系统  

Omni-directional stepping robotic system based on cellular grid map

作　　者：余翀[1] 

机构地区：[1]复旦大学信息科学与工程学院,上海200433

出　　处：《中国科学院研究生院学报》2013年第3期394-402,424,共10页Journal of the Graduate School of the Chinese Academy of Sciences

基　　金：机器人学国家重点实验室基金(R2200703)资助

摘　　要：从硬件实现和软件算法两方面,对基于蜂窝地图的全方位步进式机器人系统进行设计实现.通过对6轮全向移动平台的力学、运动学和动力学分析建模,借助基于FPGA的高精度光电里程计和激光校准装置,设计出适合蜂窝地图下运行的全方位步进式机器人.结合多维度传感器的数据融合,在蜂窝栅格地图中分2个层次,分别进行基于"六向分块"思想的局部路径规划和基于遗传算法TSP优化问题的全局路径规划.实验证明了该系统在实际运行中的精确性和鲁棒性.We design and implement an omni-directional stepping robotic system based on cellular grid map to provide physical services in intelligent space.By analyzing and modeling the mobile platform in mechanics,kinematics,and dynamics,we design the robot suitable on cellular grid map,with the help of high-precision optical odometer based on FPGA and of laser calibration device.Based on multi-dimensional sensor data fusion,we make the path planning task into two stages： six-directional local path planning and global path planning based on TSP optimization of genetic algorithm.Experiments show the accuracy and robustness of the system during the actual operation.

关 键 词：智能空间 步进式机器人 数据融合 蜂窝栅格地图 路径规划 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]