HS-SLAM: A Fast and Hybrid Strategy-Based SLAM Approach for Low-Speed Autonomous Driving

2025-05-27

AI Summary

To address the trade-off between real-time performance and accuracy in vision-inertial SLAM for low-speed autonomous driving, this paper proposes a hybrid framework integrating direct and feature-based methods. Our key contributions are: (1) a novel multi-level progressive direct method that refines IMU-preintegrated poses in a coarse-to-fine manner; (2) a dynamic keyframe selection mechanism that skips descriptor extraction for non-keyframes and discards the constant-velocity motion assumption; and (3) joint optimization of corner feature matching, adaptive keyframe selection, and IMU preintegration constraints. Evaluated on the EuRoC dataset, our method achieves higher localization accuracy than ORB-SLAM3 and improves average tracking efficiency by 15%, significantly enhancing both real-time performance and robustness.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: State Estimation
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Visual-inertial simultaneous localization and mapping (SLAM) is a key module of robotics and low-speed autonomous vehicles, which is usually limited by the high computation burden for practical applications. To this end, an innovative strategy-based hybrid framework HS-SLAM is proposed to integrate the advantages of direct and feature-based methods for fast computation without decreasing the performance. It first estimates the relative positions of consecutive frames using IMU pose estimation within the tracking thread. Then, it refines these estimates through a multi-layer direct method, which progressively corrects the relative pose from coarse to fine, ultimately achieving accurate corner-based feature matching. This approach serves as an alternative to the conventional constant-velocity tracking model. By selectively bypassing descriptor extraction for non-critical frames, HS-SLAM significantly improves the tracking speed. Experimental evaluations on the EuRoC MAV dataset demonstrate that HS-SLAM achieves higher localization accuracies than ORB-SLAM3 while improving the average tracking efficiency by 15%.

Problem

• Reducing computation burden in visual-inertial SLAM for autonomous driving
• Combining direct and feature-based methods for faster performance
• Improving tracking speed and accuracy compared to ORB-SLAM3

Innovation

• Hybrid direct and feature-based SLAM framework
• Multi-layer direct method for pose refinement
• Selective bypassing of descriptor extraction