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

Study on the Lightenning of the Weight of the Underwater Robot Mainpulator

水下机器人主动力减轻重量的研究

基本信息

批准号：
04302035
负责人：
OGAWARA Yoichi
金额：
$ 2.11万
依托单位：
KYUSYU UNIVERSITY
依托单位国家：
日本
项目类别：
Grant-in-Aid for Co-operative Research (A)
财政年份：
1992
资助国家：
日本
起止时间：
1992 至 1993
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-04302035/
关键词：
Wight Lightenning Position Force Control Underwater Robot Manipulator High Speed Learning High Trajectory Learning Feed-Forward Control Neural Network Operating Energy Minimization 海用ロボットマニピュレータ位置と力のハイブリッド制御学習制御操作エネルギー低減

项目摘要

1.Study on the force control(1)We tried to apply the Learning Feed-Forward Controller (LFFC) to the hybrid position and force control system of the underwater robot manipulator. And to hasten the learning speed, we modified the usual learning equation by adding the proportional term. As a result, the system can learn the characteristics of the controlled system with sufficiently high speed for practical use.(2)By the high speed learning type feed-forward control mode mentioned above, it is considerd to be able to construct the practical control system which is expected to control not only the position of the hand of the manipulator but also the force to the object of operation with very high adaptiveness to the nonlinearity of the manipulator dynamics and to the change of the operating condition.2.Study on the manipulator hand trajectory which minimizes the energy consumption(1)It is proved that the operating energy at the inimum consumption can be reduced to about 60% compared with that at the minimum rate of the change of acceleration of the position of the hand.(2)The trajectory of the inimum operating energy consumption abtained from the solution of the two point boundary value problem can be approximated with neural network with-sufficient accuracy for practical use. And so, it can be expected to construct the trajectory control system under the minimum operating energy consumption.As a result described above, we could establish the basic technique ;(〕SY.encircled1.〔) on the force control mode, the technique of avoiding the addition of excess force to the operating object for the lightenning of the manipulator structure,(〕SY.encircled2.〔) on the minimum operating energy consumption, the technique of the acquistion of the manipuretor hand trajectory for the lightenning of power source of the manipulator.

1.力控制研究(1)尝试将学习前馈控制器(LFFC)应用于水下机器人机械手的位置-力混合控制系统。为了加快学习速度，我们对通常的学习方程进行了修正，增加了比例项。因此，系统能够以足够高的速度学习被控系统的特性供实际使用。(2)通过上述高速学习型前馈控制模式，2.对能量消耗最小的机械手手轨迹进行了研究：(1)证明了与最小手位加速度变化率相比，最小能量消耗时的操作能量可降低60%左右；(2)最小能量消耗下的最小能量消耗；(2)最小能量消耗轨迹两点边值问题的解可用神经网络逼近，具有足够的精度，便于实际应用。因此，可以期望在最小运行能耗下构建轨迹控制系统。如上所述，我们可以建立基本技术；()SY环路1。()在力控制模式上，为了机械手结构的轻量化，避免向操作对象添加额外的力的技术，()SY环路2。()关于最小运行能耗，机械手手轨迹的获取技术，用于机械手的动力源轻量化。