AdaptSLAM: Edge-Assisted Adaptive SLAM with Resource Constraints via Uncertainty Minimization

TL;DR

AdaptSLAM is an edge-assisted SLAM system that dynamically adapts to limited communication and computation resources, minimizing uncertainty to improve localization accuracy in resource-constrained environments.

Contribution

It introduces a theoretically grounded method for selecting keyframes to optimize SLAM performance under resource constraints, integrating with ORB-SLAM3.

Findings

Reduces tracking error by 62% under constrained bandwidth

Adapts to varying resource availability effectively

Improves SLAM accuracy with resource-aware keyframe selection

Abstract

Edge computing is increasingly proposed as a solution for reducing resource consumption of mobile devices running simultaneous localization and mapping (SLAM) algorithms, with most edge-assisted SLAM systems assuming the communication resources between the mobile device and the edge server to be unlimited, or relying on heuristics to choose the information to be transmitted to the edge. This paper presents AdaptSLAM, an edge-assisted visual (V) and visual-inertial (VI) SLAM system that adapts to the available communication and computation resources, based on a theoretically grounded method we developed to select the subset of keyframes (the representative frames) for constructing the best local and global maps in the mobile device and the edge server under resource constraints. We implemented AdaptSLAM to work with the state-of-the-art open-source V- and VI-SLAM ORB-SLAM3 framework, and demonstrated that, under constrained network bandwidth, AdaptSLAM reduces the tracking error by 62% compared to the best baseline method.