Automating Human Motor Performance Ability Testing: The Case of Backward Step Detection
2017 IEEE 42nd Conference on Local Computer Networks Workshops (LCN Workshops)(2017)
摘要
Coordination under precision demands is an important aspect of human motor performance ability. It is usually evaluated through a carefully selected set of physical exercises, which follow the same scheme: An examiner presents candidates with a task to perform, monitors their correct task execution, and manually tracks the score achieved by each candidate. However, the latter two steps impose a significant cognitive load on the examiner. Even short losses of attention while carrying out the (often monotonous) test procedures may lead to erroneous results being reported. An automated solution of determining test results would not only reduce the required degree of examiner attentiveness, but potentially even accelerate testing thereby. However, such a system has not yet been proposed to the best of our knowledge. We fill this gap by presenting an approach towards the automation of the "balancing backwards" motor performance ability test in this paper. Its objective is to quantify locomotion ability by counting the number of backward steps a test subject can take when balancing on a narrow aluminum beam. We analyze the design space for backward step counting and derive a sensor configuration tailored to the autonomous detection of the number of steps taken, based on gyroscope data and laser light barriers. The system design is followed by an evaluation of its achievable accuracy levels in real-world tests, which confirm its practical viability.
更多查看译文
关键词
physical exercises,correct task execution,test procedures,automated solution,examiner attentiveness,locomotion ability,backward steps,backward step counting,backward step detection,cognitive load,backwards motor performance ability test,human motor performance ability testing,sensor configuration
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络