Look Forward to Walk Backward: Efficient Terrain Memory for Backward Locomotion with Forward Vision

TL;DR

This paper introduces LF2WB, a terrain-memory framework enabling legged robots to walk backward safely using only forward vision and proprioception, enhancing backward agility on complex terrains.

Contribution

The paper presents a novel terrain-memory system that allows backward locomotion without rearward sensors, using a delta-rule associative memory trained for efficient onboard deployment.

Findings

Improves backward agility on complex terrains

Operates with constant-time inference on low-cost hardware

Effective in both simulation and real-world tests

Abstract

Legged robots with egocentric forward-facing depth cameras can couple exteroception and proprioception to achieve robust forward agility on complex terrain. When these robots walk backward, the forward-only field of view provides no preview. Purely proprioceptive controllers can remain stable on moderate ground when moving backward but cannot fully exploit the robot's capabilities on complex terrain and must collide with obstacles. We present Look Forward to Walk Backward (LF2WB), an efficient terrain-memory locomotion framework that uses forward egocentric depth and proprioception to write a compact associative memory during forward motion and to retrieve it for collision-free backward locomotion without rearward vision. The memory backbone employs a delta-rule selective update that softly removes then writes the memory state along the active subspace. Training uses hardware-efficient parallel computation, and deployment runs recurrent, constant-time per-step inference with a constant-size state, making the approach suitable for onboard processors on low-cost robots. Experiments in both simulations and real-world scenarios demonstrate the effectiveness of our method, improving backward agility across complex terrains under limited sensing.