基于行为的多自由度机器人运动规划

提出了一种有效的基于行为的多关节机械手的运动规划。该规划将低级的运动规划和高级的行为决策结合在一起，在将位于机器人工作空间中的障碍物快速映射到姿态空间中的算法基础上，通过传感信息来计算局部障碍物的信息，从而形成姿态空间中障碍物的模型。定义了几种类型的行为来描述机械手的运动，并给出了选择行为的规则。而这些行为都是用基于姿态空间的算法可实现的。该方法适于多关节机器人在非确定环境中实时规划的要求。

Behavior based robot motion planning with many degrees of freedom

Because a robot usually operates in environments with uncertain information, path planning for the robot should be performed in real time so as to match process speed based on sensors. This paper presents an efficient approach to behavior based motion planning of a manipulator with many degrees of freedom. The method connects low level motion planning to high level behavior decision. Based on the information from sensors and an algorithm for fast mapping obstacles from a robot's workspace into configuration space, the model of configuration space obstacles is generated by computing local obstacles. In this paper, several types of behaviors are defined to formulate a manipulator's motion, while rules of choosing behaviors are also given. All the behaviors can be realized by algorithms based on configuration space. The proposed method is suitable for the requirement on real time robot path planning in uncertain environments.