IN-Sight: Interactive Navigation through Sight

TL;DR

IN-Sight is an innovative self-supervised navigation system that interacts with obstacles, using RGB-D data and a local planner to improve long-range path planning in complex environments, validated in simulation and real-world tests.

Contribution

The paper introduces IN-Sight, a novel interactive navigation framework that combines self-supervised learning, semantic mapping, and differentiable costmaps for improved obstacle navigation.

Findings

Effective long-range navigation in maze-like environments.

Successful zero-shot sim-to-real transfer on legged robot.

Enhanced obstacle interaction through end-to-end training.

Abstract

Current visual navigation systems often treat the environment as static, lacking the ability to adaptively interact with obstacles. This limitation leads to navigation failure when encountering unavoidable obstructions. In response, we introduce IN-Sight, a novel approach to self-supervised path planning, enabling more effective navigation strategies through interaction with obstacles. Utilizing RGB-D observations, IN-Sight calculates traversability scores and incorporates them into a semantic map, facilitating long-range path planning in complex, maze-like environments. To precisely navigate around obstacles, IN-Sight employs a local planner, trained imperatively on a differentiable costmap using representation learning techniques. The entire framework undergoes end-to-end training within the state-of-the-art photorealistic Intel SPEAR Simulator. We validate the effectiveness of IN-Sight through extensive benchmarking in a variety of simulated scenarios and ablation studies. Moreover, we demonstrate the system's real-world applicability with zero-shot sim-to-real transfer, deploying our planner on the legged robot platform ANYmal, showcasing its practical potential for interactive navigation in real environments.