Training toddlers seated on mobile robots to drive indoors amidst obstacles.
Training toddlers seated on mobile robots to drive indoors amidst obstacles.
复制标题
训练坐在移动机器人上的幼儿在障碍物中驾驶进入室内。
DOI：
10.1109/tnsre.2011.2114370
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
Agrawal,SunilK
中科院分区：
文献类型：
--
作者：
Chen,Xi;Ragonesi,Christina;Galloway,JamesC;Agrawal,SunilK
流动性是发展的一个因果因素。有行动障碍的儿童可能依赖力量行动来独立，因此需要先进的驾驶技能来独立运作。我们以前的研究表明，虽然婴儿可以在几个月的训练中学会使用传统的自动驾驶仪直接驾驶到目标，但他们无法在这段时间内获得驾驶时避开障碍物的高级技能。如果没有足够的驾驶训练，儿童无法安全地探索环境，其后果可能反过来增加他们发育迟缓的风险。因此，这项研究的目标是训练坐在移动的机器人上的儿童有目的地安全地在室内驾驶。在本文中，我们提出的结果，十个典型的发展幼儿训练驾驶机器人在一个障碍课程。我们还报告了一个病例研究与脊柱裂谁不能独立行走的幼儿。使用基于人工势场的算法来避开障碍物，我们在操纵杆上创建力场，训练孩子们在避开障碍物的同时导航。在这种“按需辅助”的方法中，如果儿童将操纵杆转向以期望方向为中心的力通道之外，则驾驶员的手上会受到偏置力。我们的研究结果表明，使用力反馈操纵杆可能会产生更快的学习比使用传统的操纵杆。
Mobility is a causal factor in development. Children with mobility impairments may rely upon power mobility for independence and thus require advanced driving skills to function independently. Our previous studies show that while infants can learn to drive directly to a goal using conventional joysticks in several months of training, they are unable in this timeframe to acquire the advanced skill to avoid obstacles while driving. Without adequate driving training, children are unable to explore the environment safely, the consequences of which may in turn increase their risk for developmental delay. The goal of this research therefore is to train children seated on mobile robots to purposefully and safely drive indoors. In this paper, we present results where ten typically-developing toddlers are trained to drive a robot within an obstacle course. We also report a case study with a toddler with spina-bifida who cannot independently walk. Using algorithms based on artificial potential fields to avoid obstacles, we create force field on the joystick that trains the children to navigate while avoiding obstacles. In this “assist-as-needed” approach, if the child steers the joystick outside a force tunnel centered on the desired direction, the driver experiences a bias force on the hand. Our results suggest that the use of a force-feedback joystick may yield faster learning than the use of a conventional joystick.