摘要
针对六足机器人实际运动空间相对于目标工作空间的冗余性对机器人腿部结构造成多余尺寸和能耗的问题,提出具有仿哺乳动物的三关节腿部结构的优化设计方法.分析讨论目标工作空间和实际运动空间之间的关系.通过对足端空间的运动学分析,提出基于目标工作空间的通用结构优化方法,建立不等式约束下的非线性规划的数学模型,并应用增广拉格朗日遗传算法来寻找全局优化解.通过优化,机器人的结构尺寸和关节转动范围明显减小,运动空间的冗余范围明显优于优化前.结果表明,该方法能够满足实际工作的要求并且具有较好的可操作性.而通过对于关节转矩和系统能耗的分析证明该方法能够在满足转向和越障的要求的前提下,通过减小自身关节尺寸和惯量,达到减小系统能耗的目的.
Considering the unnecessary leg dimension and energy consumption problem caused by the redundancy of the actual motion space of hexapod robot respect to the target workspace, an optimal design of the three joint structured leg with mammalian configuration was raised. The relationship between the target workspace and actual motion space was discussed. Thus, a general optimal mechanism design for the target workspace was proposed through the kinematics analysis of the foot tip. A mathematics model of nonlinear programming with inequality constraints was established, and Augmented Lagrangian Genetic Algorithm was used to find out the globally optimal solution. Through the optimization, the structure dimension and joint rotation scope is dwindled obviously and the redundancy of the motion space is improved, And the results illustrate that the optimization can satisfy the demanding of the actual tasks and is in good manipulability. At last, the joint torque and system energy consumption is analyzed, which verifled that with the requirement of turning and obstacle overcome satisfied, the total energy consumption can be reduced through decreasing the joint dimension and inertia.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2014年第5期770-776,798,共8页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金创新研究群体科学基金资助项目(51221004)
浙江省重点科技创新团队计划资助项目(2010R50036)
关键词
六足机器人
优化设计
遗传算法
工作空间
能耗
hexapod robot
optimal design
genetic algorithm
workspace
energy consumption
