Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety

Thanh Nguyen Canh, Thang Tran Viet, Thanh Tuan Tran, Ben Wei Lim

Abstract

The rise of unmanned ``dark factories'' operating without human presence demands autonomous safety systems capable of detecting and responding to multiple hazard types. We present SafeGuard ASF (Agentic Security Fleet), a comprehensive framework deploying humanoid robots for autonomous hazard detection in industrial environments. Our system integrates multi-modal perception (RGB-D imaging), a ReAct-based agentic reasoning framework, and learned locomotion policies on the Unitree G1 humanoid platform. We address three critical hazard scenarios: fire and smoke detection, abnormal temperature monitoring in pipelines, and intruder detection in restricted zones. Our perception pipeline achieves 94.2% mAP for fire or smoke detection with 127ms latency. We train multiple locomotion policies, including dance motion tracking and velocity control, using Unitree RL Lab with PPO, demonstrating stable convergence within 80,000 training iterations. We validate our system in both simulation and real-world environments, demonstrating autonomous patrol, human detection with visual perception, and obstacle avoidance capabilities. The proposed ToolOrchestra action framework enables structured decision-making through perception, reasoning, and actuation tools.

SafeGuard ASF: SR Agentic Humanoid Robot System for Autonomous Industrial Safety

Abstract

The rise of unmanned ``dark factories'' operating without human presence demands autonomous safety systems capable of detecting and responding to multiple hazard types. We present SafeGuard ASF (Agentic Security Fleet), a comprehensive framework deploying humanoid robots for autonomous hazard detection in industrial environments. Our system integrates multi-modal perception (RGB-D imaging), a ReAct-based agentic reasoning framework, and learned locomotion policies on the Unitree G1 humanoid platform. We address three critical hazard scenarios: fire and smoke detection, abnormal temperature monitoring in pipelines, and intruder detection in restricted zones. Our perception pipeline achieves 94.2% mAP for fire or smoke detection with 127ms latency. We train multiple locomotion policies, including dance motion tracking and velocity control, using Unitree RL Lab with PPO, demonstrating stable convergence within 80,000 training iterations. We validate our system in both simulation and real-world environments, demonstrating autonomous patrol, human detection with visual perception, and obstacle avoidance capabilities. The proposed ToolOrchestra action framework enables structured decision-making through perception, reasoning, and actuation tools.

Paper Structure

This paper contains 58 sections, 17 equations, 9 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Industrial Inspection Robots
  4. Fire and Hazard Detection Systems
  5. Thermal Anomaly Detection
  6. LLM/VLM for Robotics
  7. System Architecture
  8. System Overview
  9. Hardware Platform
  10. Locomotion Policy Training
  11. MDP Formulation
  12. Reward Function
  13. Training Configuration
  14. Perception Pipeline (L1)
  15. Fire and Smoke Detection
  16. ...and 43 more sections

Figures (9)

  • Figure 1: Six-layer hierarchical architecture of SafeGuard ASF. Sensor data flows through perception (L1) and understanding (L2) layers to inform decision-making (L4), which coordinates planning (L5) and action (L6) for hazard response.
  • Figure 2: Locomotion policy training pipeline using Unitree RL Lab. Training in Isaac Sim with 4096 parallel environments and domain randomization, followed by sim-to-sim validation in MuJoCo and sim-to-real deployment on the physical G1 robot.
  • Figure 3: Fire detection and response scenario. The robot detects fire via visual perception, assesses severity, coordinates suppression activation, and monitors until extinguishment. Timeline shows detection at T=0, alert at T+3s, suppression at T+18s, and confirmation at T+48s.
  • Figure 4: Qualitative results of dance training.
  • Figure 5: Qualitative results of Gangnam training.
  • ...and 4 more figures