SAFER (Safety-Aware Framework for Execution in Robotics) enables robots to complete tasks while actively identifying and mitigating safety risks.

• Employs a dedicated Safety Agent that works alongside task planning systems
• Provides real-time risk assessment throughout execution workflows
• Reduces safety violations while maintaining task completion efficiency
• Establishes concrete safety evaluation metrics for robotic operations

This research addresses a critical gap in robotic security, where traditional LLM implementations focus on task completion over safety considerations—essential for real-world deployment in sensitive environments.

Safety Aware Task Planning via Large Language Models in Robotics