SINRL: Socially Integrated Navigation with Reinforcement Learning using Spiking Neural Networks

2025-12-08

AI Summary

To address the challenge of autonomous navigation in human-robot coexistence environments, this paper proposes a hybrid spiking deep reinforcement learning (S-DRL) framework that overcomes training instability and the trade-off between energy efficiency and social compliance inherent in existing neuromorphic DRL approaches. Methodologically, the framework integrates an event-driven spiking neural network (SNN) for action generation, an artificial neural network (ANN) for value estimation, and a biologically inspired temporal feature extractor to model dynamic social interactions. Our key contributions are: (i) the first end-to-end socially aware navigation architecture unifying SNN and ANN components, significantly enhancing training stability; (ii) a measured 1.69-order-of-magnitude reduction in energy consumption while preserving navigation performance; and (iii) empirical validation of robust, energy-efficient, and socially compliant navigation in complex, high-density pedestrian scenarios.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Assisted, interactive, and conversational search
• Economics, Online Markets and Human Computation: Social networks and social learning

Abstract

Integrating autonomous mobile robots into human environments requires human-like decision-making and energy-efficient, event-based computation. Despite progress, neuromorphic methods are rarely applied to Deep Reinforcement Learning (DRL) navigation approaches due to unstable training. We address this gap with a hybrid socially integrated DRL actor-critic approach that combines Spiking Neural Networks (SNNs) in the actor with Artificial Neural Networks (ANNs) in the critic and a neuromorphic feature extractor to capture temporal crowd dynamics and human-robot interactions. Our approach enhances social navigation performance and reduces estimated energy consumption by approximately 1.69 orders of magnitude.

Problem

• Develops a hybrid DRL approach for socially integrated robot navigation
• Combines SNNs and ANNs to address unstable training in neuromorphic methods
• Reduces energy consumption while capturing temporal crowd dynamics

Innovation

• Hybrid SNN-ANN actor-critic architecture for social navigation
• Neuromorphic feature extractor captures temporal crowd dynamics
• Reduces energy consumption by ~1.69 orders of magnitude