ATTENTION-BASED MAP ENCODING FOR GENERALIZED ROBOT LOCOMOTION

Number: 20260169493
Published: 2026-06-18
Filed: 2025-12-11
Assignee: DISNEY ENTERPRISES, INC.
Inventors: GRANDIA; Ruben Jelle, ZHANG; Chong, HE; Junzhe, HUTTER; Marco, JENELTEN; Fabian Pasqual, BÄCHER; Moritz Niklaus
CPC: G05D1/646; G06N3/006; G06N3/008; G06N3/044; G06N3/045; G06N3/0464; G06N3/08; G06N3/092
Verdict: High Hardware
First reported: 2026-W29 (2026-07-15)
Source: Google Patents · FreePatentsOnline

The keeper's note

The present invention sets forth techniques for generating robot action commands directed to robot locomotion.

Abstract

The present invention sets forth techniques for generating robot action commands directed to robot locomotion. The techniques include encoding map scans representing terrain and proprioception data representing a current state of a robot. The techniques also include generating point-wise map features based on the encoded map scan, and transmitting the map features to a multi-head attention (MHA) mechanism. The MHA mechanism assigns an attention weight to each of multiple terrain points included in the map features, conditioned on the proprioception data. The MHA mechanism generates a map encoding representing the terrain points and their corresponding attention weights, where higher attention weights are associated with terrain points that may provide suitable footholds for the robot. Based on a concatenation of the map encoding and the proprioception data, a multilayer perceptron (MLP) generates robot action commands, such as joint movement commands, causing the robot to traverse the terrain.

Background

BACKGROUND Field of the Various Embodiments

Embodiments of the present disclosure relate generally to machine learning and robot locomotion and, more specifically, to techniques for implementing attention-based map encoding for generalized robot locomotion. Description of the Related Art

In the field of mobile robots, legged robots may be better suited to navigating rough, uneven, broken, or otherwise irregular terrain compared to other types of mobile robots such as wheeled or tracked robots. Further, legged robots may cause less damage or erosion to terrain compared to wheeled or tracked robots.

Successful terrain navigation requires that a legged robot calculate suitable locations within the terrain to place its feet. A suitable location for foot placement, or foothold, may include a location that will enable the robot to progress toward a specified goal while also enabling the robot to maintain and/or recover its balance during locomotion.

Existing techniques for robot locomotion may include one or more machine learning models trained in an end-to-end manner using a deep reinforcement learning (DRL) technique. While these methods may demonstrate adequate robustness against uncertainty and noise included in the model inputs, one drawback of these techniques is that the models may struggle to identify valid footholds on sparse terrain or learn from the identified footholds.

Existing techniques may include other machine learning model-based

Claims

1. A computer-implemented method for generating robot action commands, the computer-implemented method comprising: generating embedded map features based at least on a three-dimensional (3D) representation of a terrain; generating a query based at least on robot proprioception data; generating, using multi-head attention and based at least on the query and the embedded map features, attention weights associated with a plurality of terrain points included in the terrain; and generating one or more robot action commands based at least on the attention weights and the robot proprioception data. || 13. One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of: generating embedded map features based at least on a three-dimensional (3D) representation of a terrain; generating a query based at least on robot proprioception data; generating, using multi-head attention and based at least on the query and the embedded map features, attention weights associated with a plurality of terrain points included in the terrain; and generating one or more robot action commands based at least on the attention weights and the robot proprioception data. || 18. A system comprising: one or more memories storing instructions; and one or more processors for executing the instructions to: generate embedded map features based at least on a three-dimensional (3D) representation of a terrain; generate a query based at least on robot proprioception data; generate, using multi-head attention and based at least on the query and the embedded map features, attention weights associated with a plurality of terrain points included in the terrain; and generate one or more robot action commands based at least on the attention weights and the robot proprioception data.