Research map / Multi-agent systems and games / Human-AI and goal-conditioned
SOCIALGYM 2.0: Simulator for Multi-Robot Learning and Navigation in Shared Human Spaces
Rohan Chandra, Zayne Sprague, Joydeep Biswas
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Presents SocialGym 2.0, an open-source multi-agent simulator for social robot navigation in constrained indoor scenarios (doorways, hallways, intersections, roundabouts) with MARL support via PettingZoo and Stable Baselines3.
Problem
Existing navigation simulators focus on single robots in open spaces with simple environments and limited configurability, so they do not support multi-robot social navigation learning in complex shared human spaces.
Approach
A configuration-over-convention simulator that exposes multi-agent RL through PettingZoo/Stable Baselines3 APIs, with customizable observations, rewards, kinodynamic robot models, 2D maps/navigation graphs, and social mini-game scenarios plus social navigation metrics.
Limitations
- CPU resource constraints
- lack of multi-threading and parallel processing optimization
- only discrete actions; no state-of-the-art MARL algorithms such as QMIX, MAPPO, MADDPG yet
Experimental setup
Metrics: partial and full success rates velocity changes average stopping time collision rates
Baselines: SocNavBench SEAN2.0 CrowdBOT
Builds on: PettingZoo Stable Baselines3 SocialGym
Closest papers by meaning
- Following the Human Thread in Social Navigation (ICLR 2025 · similarity 0.88)
- Habitat 3.0: A Co-Habitat for Humans, Avatars, and Robots (ICLR 2024 · similarity 0.87)
- POGEMA: A Benchmark Platform for Cooperative Multi-Agent Pathfinding (ICLR 2025 · similarity 0.85)
- ALSO: Adversarial Online Strategy Optimization for Social Agents (ICML 2026 · similarity 0.85)
- TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios (ICML 2026 · similarity 0.85)
- LocoVR: Multiuser Indoor Locomotion Dataset in Virtual Reality (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
