Simultaneous Localization and Mapping Algorithm Based on the Asynchronous Fusion of Laser and Vision Sensors

Area of Science:

• Robotics
• Computer Vision
• Sensor Fusion

Background:

• Simultaneous Localization and Mapping (SLAM) is crucial for autonomous robot navigation.
• Existing SLAM algorithms often face challenges with sensor data synchronization and redundancy.
• Assistant robots require robust and accurate real-time pose estimation.

Purpose of the Study:

• To propose a novel weighted asynchronous fusion algorithm for SLAM in assistant robots.
• To improve tracking accuracy and operational efficiency compared to synchronous fusion methods.
• To enhance the robustness of SLAM in degraded environmental conditions.

Main Methods:

• Developed a SLAM algorithm using weighted asynchronous fusion of laser and vision sensors.
• Utilized attitude estimation from visual sensors as a prior for laser sensor attitude estimation.
• Introduced an angle-based weighting coefficient to improve measurement confidence based on robot running state.

Main Results:

• The proposed asynchronous fusion algorithm demonstrated improved tracking accuracy.
• The algorithm showed faster operation speed and higher pose estimation frequency.
• Experimental results confirmed the algorithm's robustness and effectiveness in degraded environments.

Conclusions:

• Weighted asynchronous sensor fusion offers significant advantages for robot SLAM.
• The algorithm provides more accurate priors, faster processing, and higher accuracy than synchronous methods.
• This approach enhances the reliability and performance of assistant robots in complex environments.