Decision-Making Authority, Team Efficiency and Human Worker Satisfaction in Mixed Human-Robot Teams

Abstract

Abstract-In manufacturing, advanced robotic technology has opened up the possibility of integrating highly autonomous mobile robots into human teams. However, with this capability comes the issue of how to maximize both team efficiency and the desire of human team members to work with robotic counterparts. We hypothesized that giving workers partial decision-making authority over a task allocation process for the scheduling of work would achieve such a maximization, and conducted an experiment on human subjects to test this hypothesis. We found that an autonomous robot can outperform a worker in the execution of part or all of the task allocation (p < 0.001 for both). However, rather than finding an ideal balance of control authority to maximize worker satisfaction, we observed that workers preferred to give control authority to the robot (p < 0.001). Our results indicate that workers prefer to be part of an efficient team rather than have a role in the scheduling process, if maintaining such a role decreases their efficiency. These results provide guidance for the successful introduction of semi-autonomous robots into human teams.

Efficient Model Learning from Joint-Action Demonstrations for Human-Robot Collaborative Tasks

“…We posited that the robot anticipatory actions of the learned MOMDP policy, as well as of the MOMDP policy from the hand-coded reward, would result in faster task execution and better team fluency compared with manually annotating robot actions. Automating robot behaviors [11] and, in particular, enabling anticipatory actions [12] has previously resulted in significant improvements in team efficiency and fluency in manufacturing applications.…”

Observer Perception of Dominance and Mirroring Behavior in Human-Robot Relationships

“…This is in light of the fact that people who collaborate with robots may need to be in subordinate roles (or even prefer them, cf. [19]), especially if robots have superior abilities in the task at hand.…”

“…Evaluation acts have previously been used to establish dominance (e.g., [34]). Work by Gombolay et al [19] suggested that robots with greater decisionmaking authority are preferred over those with less authority, although they did not assess the attitudes of observers. We hypothesize that a dominant robot will be viewed as less trustworthy, agreeable and attractive than a submissive one.…”

An ontology to enable optimized task partitioning in human-robot collaboration for warehouse kitting operations

“…Thus, Equation (14) enforces that every atomic task for an object is at least attempted once by the agents with the maximum number of trials by all the agents set by the right-hand side vector in Equation (15). Equation (16) ensures that an agent does not begin more than one object task at any time instant.…”

“…Hence, we develop a method to optimally partition tasks between humans and robots in hybrid work cells using kitting, which involves packing of individually separate but related parts into different units, as the example application. While optimal task assignment in human-robot teams [12][13][14] has been studied for assembly operations, we focus on generating higherlevel task partitioning policies for agent (robot or human) groups instead of lower-level task allocation schedules for the individual agents. The optimization method makes use of an ontological representation of the tasks that builds on a large body of literature on developing standards 15-17 for defining robot task ontology, and applying such ontologies for service robots, 18-20 surgical robots, 21 and even industrial robots.…”