Collaboration of human pickers and crop-transporting robots during harvesting – Part I: Model and simulator development

摘要: Abstract Some specialty crops, such as strawberries and table grapes, are harvested by large crews of pickers who spend significant amounts time carrying empty full (with the crop) trays. A step toward increasing harvest automation for crops is to deploy harvest-aid robots that transport trays, thus efficiency reducing pickers’ non-productive walking times. To end, this work addresses human-robot collaboration modeling in a harvesting context. First, framework all-manual robot-aided was developed, which can be used off-line simulation system designers, but also representation model robot control, during real-time operation. serve both functions, utilizes hybrid systems picker activities. Finite state machines discrete operating states, difference equations describe motion mass transfer within each state. capture variability human behavior performance harvesting, activity stochastic parameters (e.g., picking time, speed) estimated measurements harvesting. The does not require direct yield measurements, available most crops. Second, simulator developed based on model. For given field crew size, samples all generate many instances operation, estimates metrics operation efficiency. Part II presents calibration evaluation data, case study evaluates effect various scheduling algorithms