Date of Graduation
This study evaluates factors that could affect human performance in the event of an unexpected robot motion. The study evaluated the following factors associated with a robot work station: illumination level, noise level, the luminance contrast between the robot and its background, the amount of attention required by a task being performed while monitoring the movement of the robot, the motion speed of the robot, and finally the field where unexpected motions were initiated. The dependent variable measured was the distance the robot moved before the subject detected and responded to the unexpected motion. In a main experiment, nine subjects with different occupations (5 students & 4 industrial workers) participated. The results demonstrated many things. First, robot motion speed and task demand had an important impact on robot overrun distance. A longer overrun distance was observed with both increasing robot motion speed or more complex task demand. Second, illumination affected subject's performance; a longer overrun distance was observed with a low illumination level (10 Lux). Overrun distance, however, remained unaffected with increasing illumination level above 100 Lux. Furthermore, unexpected robot motions initiated in the peripheral field resulted in a longer overrun distance than motions initiated in the central field. The analysis of the robot motion speed by illumination interaction suggested that a low illumination level may further increase overrun distance with increasing robot motion speed. The robot motion speed by luminance contrast interaction also suggested that when increasing robot motion speed, the overrun distance increases with low background-to-robot arm luminance contrast ratios. A multiple regression model was also derived for predicting robot overrun distance. Based on the assumption that a person would not approach a robot closer to a clear human vision distance of 20-30 cm, the model suggested that a programming speed of 17 cm/sec provides a greater safety margin for contact avoidance compared to the American National Standard Institute (ANSI) recommended speed (25 cm/sec), and the 17 cm/sec should be recommended as the maximum robot slow speed for work conducted in the vicinity of an industrial robot not equipped with dead-man switches.
Beauchamp, Yves, "A study of human performance in the event of an unexpected robot motion." (1989). Graduate Theses, Dissertations, and Problem Reports. 8458.