CREATE: Cross-Layer Resilience Characterization and Optimization for Efficient yet Reliable Embodied AI Systems

2026-01-20

AI Summary

This work addresses the challenges of deploying embodied AI systems on resource-constrained edge devices, where high energy consumption and voltage scaling compromise reliability. The authors propose CREATE, a cross-layer co-optimization framework that, for the first time, systematically characterizes the heterogeneous fault tolerance properties of embodied AI across abstraction layers. CREATE integrates circuit-level anomaly detection, LLM weight rotation for enhanced robustness, and a hardware-software co-designed adaptive voltage scaling mechanism. Evaluated under real-world conditions, the approach achieves a 40.6% reduction in average computational energy compared to nominal-voltage baselines, while decreasing chip-level energy consumption by 29.5%–37.3% and extending battery life by 15%–30%, all without compromising task success rates.

Technology Category

• Intelligent Robots: Embodied AI
• Machine Learning: Hardware-aware ML
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Embodied Artificial Intelligence (AI) has recently attracted significant attention as it bridges AI with the physical world. Modern embodied AI systems often combine a Large Language Model (LLM)-based planner for high-level task planning and a reinforcement learning (RL)-based controller for low-level action generation, enabling embodied agents to tackle complex tasks in real-world environments. However, deploying embodied agents remains challenging due to their high computation requirements, especially for battery-powered local devices. Although techniques like lowering operating voltage can improve energy efficiency, they can introduce bit errors and result in task failures. In this work, we propose CREATE, a general design principle that leverages heterogeneous resilience at different layers for synergistic energy-reliability co-optimization. For the first time, we conduct a comprehensive error injection study on modern embodied AI systems and observe an inherent but heterogeneous fault tolerance. Building upon these insights, we develop an anomaly detection and clearance mechanism at the circuit level to eliminate outlier errors. At the model level, we propose a weight-rotation-enhanced planning algorithm to improve the fault tolerance of the LLM-based planner. Furthermore, we introduce an application-level technique, autonomy-adaptive voltage scaling, to dynamically adjust the operating voltage of the controllers. The voltage scaling circuit is co-designed to enable online voltage adjustment. Extensive experiments demonstrate that without compromising task quality, CREATE achieves 40.6% computational energy savings on average over nominal-voltage baselines and 35.0% over prior-art techniques. This further leads to 29.5% to 37.3% chip-level energy savings and approximately a 15% to 30% improvement in battery life.

Problem

• Embodied AI
• Energy Efficiency
• Reliability
• Voltage Scaling
• Fault Tolerance

Innovation

• cross-layer resilience
• embodied AI
• fault tolerance
• adaptive voltage scaling
• energy-reliability co-optimization