权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Dynamics and Control of Robotic Manipulation

机器人操纵动力学与控制

基本信息

批准号：
63550192
负责人：
YAMASHITA Tadashi
金额：
$ 1.41万
依托单位：
Kyushu Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for General Scientific Research (C)
财政年份：
1988
资助国家：
日本
起止时间：
1988 至 1990
项目状态：
已结题

项目摘要

The object of this project is to increase the manipulative capabilities of robotic systems so that they can be applied to various fields demanding higher functions beyond those possessed by industrial robots of present conventional type. The following three subjects were studied during the research period : Control of basic operations in the manipulation ; aiding methods for remote manipulations ; application of manipulation concepts to locomotion. Main results of each subject are the following.1. Control of Manipulation Fuzzy control and sliding mode control were applied to generate a specified grasping force. In the fuzzy control experiment, both direct and indirect reasoning gave satisfactory results by proper tuning of controller parameters in experiments done by a commercially available sliding robot. An effectiveness of a learning algorithm to adjust the parameter values was demonstrated, too. In the sliding mode control experiments a new type of a finger mechanism actuated by rubber artificial muscles used. The dynamic characteristics of the mechanism were determined experimentally. Results of control experiments were fair. The following experiments were, also, carried out : an adaptive control of a flexible arm ; hybrid control of inserting a peg into hole.2. Operator Aiding in Remote Manipulation Experiments were done by using a simulator by ourselves. Effectiveness of aids by computer in the final approaching stage to an object were shown : decrease of both misoperation and task time.3. Unified Approach to Manipulation and Locomotion The concept of manipulability that is an effective measure of the mobility capability of a manipulator was successfully applied to explain the transient characteristics of the body motion in the initial stage of the human walking. The present resech trend related to the unified approach.

该项目的目标是提高机器人系统的操纵能力，使其能够应用于要求比目前传统类型的工业机器人具有更高功能的各种领域。在研究期间，研究了以下三个主题：操纵中基本操作的控制;远程操纵的辅助方法;操纵概念在运动中的应用。主要研究结果如下：1.机械手的控制采用模糊控制和滑模变结构控制来产生指定的抓取力。在模糊控制实验中，直接和间接的推理得到了满意的结果，通过适当的调整控制器参数在实验中所做的商业上可用的滑动机器人。一个学习算法来调整参数值的有效性也被证明。在滑模控制实验中，采用了一种新型的橡胶人工肌肉驱动手指机构。通过实验确定了该机构的动态特性。对照实验结果一般。并进行了以下实验：柔性臂的自适应控制;插桩入孔的混合控制.操作员辅助远程操作的实验是通过使用自己的模拟器。在最后接近目标的阶段，计算机辅助的效果是：减少了误操作和任务时间。操作与运动的统一方法可操作性是衡量机械手移动能力的一个有效指标，它被成功地应用于解释人体在行走初始阶段的瞬态运动特性。目前的研究趋势与统一的方法有关。