【作者】 郭璐；

【导师】 沈模卫；

【作者基本信息】 浙江大学 ， 应用心理学， 2006， 博士

【摘要】 上肢伸展姿势及运动轨迹预测是工效学领域的前沿课题。目前已有的上肢伸展姿势及运动轨迹预测方法在预测准确性、自然性和实时性方面尚存在缺陷。本研究首先根据人体上肢和躯干的生理特点,对其进行了合理建模,并将站姿人体右侧上肢工作空间均匀划分为600个子区域,通过实时测量操作者执行上肢伸展运动的运动轨迹,引入曲线拟合和基于自组织映射理论的权重分布核函数相结合的方法,分别实现了站姿人体躯干非受限条件下,以典型定位运动(研究一)和快速定位运动(研究二)速度完成点对点指向任务、以典型定位运动速度完成按键任务(研究三)时,右手食指及达工作空间内任意目标位置的上肢伸展姿势及运动轨迹预测。整合上述结果,建立了基于任务的上肢伸展姿势及运动轨迹预测系统。该系统可直接应用于工效学领域计算机三维仿真人体的建立和发展,同时可直接应用于对航空航天座舱、核电站中央控制室、交通工具驾驶舱等的工效学设计与评价。本研究获得了如下主要结论:(1)执行上肢伸展运动时,上肢各关节对伸展运动所起的作用随时间而变化,且同一关节所含各自由度对伸展运动的贡献也存在差异。(2)上肢伸展姿势及运动轨迹主要由及达目标所处的空间位置决定,但同时受伸展运动速度和控制器类型的影响。(3)将曲线拟合和基于自组织映射理论的权重分布核函数相结合的方法,不仅适用于上肢伸展姿势及运动轨迹的预测,同时可为其他肢体部分运动轨迹的预测提供指导和借鉴作用。(4)本研究所建立的基于任务的上肢伸展姿势及运动轨迹预测系统可满足工效学领域上肢伸展姿势及运动轨迹预测的准确性、自然性和实时性需要。更多还原

【Abstract】 Upper body reach posture and trajectories prediction has been one of the focus topics. in the field of Human Factors. The existing predicting methods, however, should be improved in prediction accuracy, the attaiment of natural posture or trajectories and the real time of one prediction. The present study has adopted new perspectives into this field. In the first, the right arm and torso were specified as an eleven-degrees-of freedom kinematic linkage according to the human body physiological features. Secondly, the workplace was divided equally into 600 regions. Thirdly, based on real-time longitudinal data collected through three-dimensional motion capture system during the period of the right hand reaching 600 targets in the workplace, the fitting-curve method and the kernel function which stems from self-organization map theory were introduced to predict upper body reach posture and trjectories. The prediction was made in three conditions namely pointing the target button by natural speed Research i, pointing the botton by high speed Research ii and pushing the target button by natural speed Research iii . On the basis of those results, an upper body reach posture and trajectories prediction computer system was established. This system can be used in the simulation of upper body motion in computer three-dimensional human modeling, and also in ergonomic design evluation.The main findings are as follows:(1) Different parts of the right arm play different roles during the reach motion period, even the actions in different degrees-of-freedom of the same articulation are distinct.(2) The reach posture and trajectories mainly depend on the target position, however they are also affected by the motion speed and the control type.(3) The approach which integrating the workspace, curving fitting method and kernel functions can be used in upper body reach posture and trajectoties prdiction, and also can be introduced into modeling the motion of other body parts.(4) The established upper body reach posture and trajectories prediction computer system can satisfy the ergonomics design and evaluation, because of its accuracy, naturalness and real-time prediciton.更多还原