Breaking the Human-Robot Deadlock: Surpassing Shared Control Performance Limits with Sparse Human-Robot Interaction

Abstract

Abstract-Human machine teaming has, for decades, been conceptualized as a function allocation (FA) or levels of autonomy (LOA) process: the human is suited for some tasks, while the machine is suitable for others, and as machines improve they take over duties previously assigned to humans. A wide variety of methods-including adaptive, adjustable, blended, supervisory and mixed initiative control, implemented discretely or continuously, as potential fields, as virtual fixture interfaces, or haptic interfaces-are derivatives of FA/LOA. We formalize FA/LOA (and all their derivatives) under a single mathematical formulation called classical shared control (CSC). Despite the widespread adoption of CSC, we prove that it fails to optimize human and robot agreement and intent if either the human or robot model displays "intention ambiguity" (e.g., the human's intended goal is unclear or the robot finds multiple viable solutions). Practically, this suboptimality can manifest as unnecessary and unresolvable disagreement (an unnecessary deadlock). For instance, if the robot chooses to go left around an obstacle and the human chooses to go right, CSC only provides two solutions: freeze in place or collide with the obstacle (we provide a wide variety of failure examples in [52], https://arxiv.org/abs/1611.09490). We find that CSC suboptimality stems from arbitrating over model samples, rather than over models. Our key insight is thus to arbitrate over human and robot distributions; we prove this method optimizes human and robot agreement and intent and resolves deadlocking. Our key contribution is computationally efficient distribution arbitration: if the human and robot carry N our joint has fewer modes than the individual agent models. We call our approach N min -sparse generalized shared control.

Generalized Shared Control versus Classical Shared Control: Illustrative Examples

“…In [13], we proved that CSC, despite its widespread usage, is only suitable as a decision fusion architecture for simple scenarios-in particular, CSC is inappropriate for scenarios displaying multimodality 1 because it can provoke unnecessary "disagreement" between the operator and the autonomy. For instance, if an operator is trying to navigate a remote platform through a dynamic arena (e.g., search and rescue robots operating in urban areas, military robots deployed into hostile environments, or even a telepresence robot trying to navigate a crowded office), CSC is nearly guaranteed to induce unnecessary operatory-autonomy disagreement.…”

“…Shared control fuses operator inputs and autonomy inputs into a single command. However, if environmental or operator predictions are multimodal, state of the art approaches are suboptimal with respect to safety, efficiency, and operator-autonomy agreement: even under mildly challenging conditions, existing approaches can fuse two safe inputs into an unsafe shared control [13]. Multimodal conditions are common to many real world applications, such as search and rescue robots navigating disaster zones, teleoperated robots facing communication degradation, and assistive driving technologies.…”

“…Multimodal conditions are common to many real world applications, such as search and rescue robots navigating disaster zones, teleoperated robots facing communication degradation, and assistive driving technologies. In [11,13], we introduced a novel approach called generalized shared control (GSC) that simultaneously optimizes autonomy objectives (e.g., safety and efficiency) and operator-autonomy agreement under multimodal conditions; this optimality prevents such unsafe shared control. In this paper, we describe those results in more user friendly language by using illustrations and text.…”

“…As argued in [3], an approach called linear blending (weighted averaging of the human and machine inputs) is the de-facto decision fusion architecture in many implementations of the above application spaces. More generally, [13] argues that "human machine teaming has, for decades, been conceptualized as a function allocation (FA) or levels of autonomy (LOA) process: the human is suited for some tasks, while the machine is suitable for others, and as machines improve they take over duties previously assigned to humans. A wide variety of methods-including adaptive, adjustable, blended, supervisory and mixed initiative control, implemented discretely or continuously, as potential fields, as virtual fixture interfaces, or haptic interfaces-are derivatives of FA/LOA."…”

Review of Advanced Medical Telerobots

“…Authors conclude that this is a promising approach that still requires investigation to become more effective. Other similar shared autonomy formulations can be found in [355,356]. We can also cite the works of Zakerimanesh et al, that permit multilateral teleoperation (using nonlinear authority blending) for remote applications featuring time-varying communication delays, actuator saturations, nonlinearity in the dynamics (which corresponds to a common teleoperation practical situation), and more particularly for follower robots with redundancy, which is interesting with medical robots as this redundancy is typically used to avoid collisions with the staff and other devices around the patient.…”