vAccSOL: Efficient and Transparent AI Vision Offloading for Mobile Robots

2026-03-17

AI Summary

Mobile robots struggle to efficiently execute modern AI vision tasks due to limited computational resources, energy constraints, and the closed ecosystems of embedded accelerators. To address this challenge, this work proposes vAccSOL, a framework that leverages the neural network compiler SOL to generate lightweight inference libraries and integrates with the runtime system vAccel to enable transparent heterogeneous scheduling without requiring modifications to application code. This approach dynamically partitions vision workloads between on-robot and edge nodes, overcoming the limitations of proprietary software stacks while balancing hardware-specific optimizations with deployment flexibility. Experimental evaluation on a real quadruped robot demonstrates that, compared to a PyTorch baseline, vAccSOL reduces power consumption by 80% on the robot and 60% at the edge, while achieving up to a 24× increase in vision pipeline throughput.

Technology Category

• Computer Vision: Vision for Robotics & Autonomous Driving
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Hardware-aware ML

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Mobile robots are increasingly deployed for inspection, patrol, and search-and-rescue operations, relying on computer vision for perception, navigation, and autonomous decision-making. However, executing modern vision workloads onboard is challenging due to limited compute resources and strict energy constraints. While some platforms include embedded accelerators, these are typically tied to proprietary software stacks, leaving user-defined workloads to run on resource-constrained companion computers. We present vAccSOL, a framework for efficient and transparent execution of AI-based vision workloads across heterogeneous robotic and edge platforms. vAccSOL integrates two components: SOL, a neural network compiler that generates optimized inference libraries with minimal runtime dependencies, and vAccel, a lightweight execution framework that transparently dispatches inference locally on the robot or to nearby edge infrastructure. This combination enables hardware-optimized inference and flexible execution placement without requiring modifications to robot applications. We evaluate vAccSOL on a real-world testbed with a commercial quadruped robot and twelve deep learning models covering image classification, video classification, and semantic segmentation. Compared to a PyTorch compiler baseline, SOL achieves comparable or better inference performance. With edge offloading, vAccSOL reduces robot-side power consumption by up to 80% and edge-side power by up to 60% compared to PyTorch, while increasing vision pipeline frame rate by up to 24x, extending the operating lifetime of battery-powered robots.

Problem

• mobile robots
• vision offloading
• resource constraints
• energy efficiency
• heterogeneous platforms

Innovation

• vision offloading
• neural network compiler
• heterogeneous execution
• edge computing
• mobile robotics