Competence-Aware Path Planning via Introspective Perception

TL;DR

This paper introduces CPIP, a Bayesian, model-free framework enabling robots to learn and anticipate perception-related failures for improved path planning in new environments, without prior failure mode enumeration.

Contribution

The paper presents a novel, structured, model-free approach for competence-aware path planning that learns perception errors and failure prediction without prior failure enumeration or location-specific data.

Findings

CPIP outperforms baseline methods in simulation tasks.

CPIP effectively predicts perception failures in real robot experiments.

The approach adapts to perceptually challenging environments.

Abstract

Robots deployed in the real world over extended periods of time need to reason about unexpected failures, learn to predict them, and to proactively take actions to avoid future failures. Existing approaches for competence-aware planning are either model-based, requiring explicit enumeration of known failure modes, or purely statistical, using state- and location-specific failure statistics to infer competence. We instead propose a structured model-free approach to competence-aware planning by reasoning about plan execution failures due to errors in perception, without requiring a priori enumeration of failure sources or requiring location-specific failure statistics. We introduce competence-aware path planning via introspective perception (CPIP), a Bayesian framework to iteratively learn and exploit task-level competence in novel deployment environments. CPIP factorizes the competence-aware planning problem into two components. First, perception errors are learned in a model-free and location-agnostic setting via introspective perception prior to deployment in novel environments. Second, during actual deployments, the prediction of task-level failures is learned in a context-aware setting. Experiments in a simulation show that the proposed CPIP approach outperforms the frequentist baseline in multiple mobile robot tasks, and is further validated via real robot experiments in an environment with perceptually challenging obstacles and terrain.