An Energy-Efficient and Runtime-Reconfigurable FPGA-Based Accelerator for Robotic Localization Systems

TL;DR

This paper introduces an FPGA-based accelerator for robotic localization that is energy-efficient, reconfigurable at runtime, and achieves over five times the performance of existing solutions by leveraging SLAM-specific optimizations.

Contribution

It presents a novel FPGA-based design that is both energy-efficient and dynamically reconfigurable during operation for improved robotic localization.

Findings

Over 5x performance improvement over state-of-the-art

Energy savings through runtime reconfiguration

Maintains accuracy and performance across environments

Abstract

Simultaneous Localization and Mapping (SLAM) estimates agents' trajectories and constructs maps, and localization is a fundamental kernel in autonomous machines at all computing scales, from drones, AR, VR to self-driving cars. In this work, we present an energy-efficient and runtime-reconfigurable FPGA-based accelerator for robotic localization. We exploit SLAM-specific data locality, sparsity, reuse, and parallelism, and achieve >5x performance improvement over the state-of-the-art. Especially, our design is reconfigurable at runtime according to the environment to save power while sustaining accuracy and performance.