Review on SLAM of Mobile Robot for Processing Inspection Based on Multi-Sensor Fusion

Chenghuang Ye; Ruiyou Huang; Xiubao Zheng; Xiaoyu Feng; Zhangyu Zhang

doi:10.62051/ijmee.v8n3.03

Authors

Chenghuang Ye
Ruiyou Huang
Xiubao Zheng
Xiaoyu Feng
Zhangyu Zhang

DOI:

https://doi.org/10.62051/ijmee.v8n3.03

Keywords:

Multi-sensor Fusion, SLAM, Machining Inspection, Mobile Robot, ROS, Path Planning

Abstract

With the rapid advancement of Industry 4.0 and smart manufacturing, traditional fixed processing and inspection equipment can no longer meet the flexible manufacturing demands of large-scale and complex-shaped components. Mobile processing and inspection robots, with their high adaptability and efficiency, have become critical equipment in modern workshops. This paper focuses on analyzing two mainstream solutions—visual-inertial fusion (VI-SLAM) and laser-visual fusion—emphasizing the advantages of multi-sensor fusion in enhancing robustness and precision. It provides a detailed exploration of the visual SLAM framework and algorithmic advancements based on feature points and deep learning. Additionally, the study investigates the collaborative design of global path planning and local obstacle avoidance algorithms, aiming to offer technical references for reliable autonomous navigation of mobile robots to preset workpoints in known environments.

References

[1] Huang BC, Zhang M, Li H. A review of SLAM multi-sensor calibration techniques [J]. Journal of Automation Chemistry, 2022,48(5):1023-1038.

[2] DAVISON A J, REID I D, MOLTON N D, et al. MonoSLAM: Real-time single camera SLAM[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(6): 1052-1067.

[3] LI B, YANG C, WANG L, et al. DXSLAM: A robust and efficient visual SLAM system with deep features[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 12345-12354.

[4] TIWARI S, MISHRA A, MURTY M N. pRGBD-SLAM: A joint framework for monocular SLAM and depth prediction using narrow and wide baseline constraints[J]. IEEE Transactions on Robotics, 2021, 37(2): 563-578.