RoboLayout: Differentiable 3D Scene Generation for Embodied Agents

TL;DR

RoboLayout advances 3D scene generation by integrating agent-aware reasoning and reachability constraints, producing navigable, feasible layouts for diverse embodied agents in indoor environments.

Contribution

It introduces agent-aware reasoning and explicit reachability constraints into differentiable layout optimization, enabling environment design tailored to various embodied agents.

Findings

Generates navigable and feasible indoor scene layouts for diverse agents.

Improves optimization stability and convergence efficiency.

Maintains semantic alignment and physical plausibility.

Abstract

Recent advances in vision language models (VLMs) have shown strong potential for spatial reasoning and 3D scene layout generation from open-ended language instructions. However, generating layouts that are not only semantically coherent but also feasible for interaction by embodied agents remains challenging, particularly in physically constrained indoor environments. In this paper, RoboLayout is introduced as an extension of LayoutVLM that augments the original framework with agent-aware reasoning and improved optimization stability. RoboLayout integrates explicit reachability constraints into a differentiable layout optimization process, enabling the generation of layouts that are navigable and actionable by embodied agents. Importantly, the agent abstraction is not limited to a specific robot platform and can represent diverse entities with distinct physical capabilities, such as service robots, warehouse robots, humans of different age groups, or animals, allowing environment design to be tailored to the intended agent. In addition, a local refinement stage is proposed that selectively reoptimizes problematic object placements while keeping the remainder of the scene fixed, improving convergence efficiency without increasing global optimization iterations. Overall, RoboLayout preserves the strong semantic alignment and physical plausibility of LayoutVLM while enhancing applicability to agent-centric indoor scene generation, as demonstrated by experimental results across diverse scene configurations.