Implicit Traffic Signals: A Systems Approach to Human-Robot Navigation

Robots are complex hardware/software systems. What is often neglected when building robot systems is that humans are part of the system, too. If we are to have robots that operate as peers alongside people, then robots need a new mix of functional and social skills. Human cooperation is facilitated by interfaces that support compositional reasoning and allow people to perform sophisticated tasks together extemporaneously. Performing human-robot joint computation requires assorted mutual information that people routinely deploy for ordinary collaborative tasks. Collaboration with people also requires robots to solve consensus, synchronization, and resource management the way people do. Much of this collaboration often happens implicitly. I illustrate how these concepts interact in the application of social navigation, which I argue is a first-class example of teamwork. In this setting, human and robot participants avoid collision by legibly conveying intended passing sides via nonverbal cues like path shape. A topological representation using the braid group enables the robot to reason about a small enumerable set of passing outcomes. I show how implicit communication of topological group plans achieves rapid convergence to a team consensus, and how a robot in the team can deliberately influence the ultimate outcome to maximize joint performance, yielding human comfort with the robot.