• Laser & Optoelectronics Progress
  • Vol. 61, Issue 8, 0811001 (2024)
Longyun Zhao1,2, Hongjun San1,2,*, Jiupeng Chen1,2, and Zhen Peng1,2
Author Affiliations
  • 1Faculty of Mechanical and Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, Yunnan , China
  • 2Key Laboratory of Intelligent Manufacturing Technology for Advanced Equipment in Yunnan Province, Kunming 650500, Yunnan , China
  • show less
    DOI: 10.3788/LOP231307 Cite this Article Set citation alerts
    Longyun Zhao, Hongjun San, Jiupeng Chen, Zhen Peng. Laser Location of Mobile Robot Based on ICP Algorithm[J]. Laser & Optoelectronics Progress, 2024, 61(8): 0811001 Copy Citation Text show less
    Block diagram of location method
    Fig. 1. Block diagram of location method
    Schematic of multi-resolution search
    Fig. 2. Schematic of multi-resolution search
    ICP algorithm principle
    Fig. 3. ICP algorithm principle
    Submap construction
    Fig. 4. Submap construction
    Schematic of multi-point cloud density matching
    Fig. 5. Schematic of multi-point cloud density matching
    Diagram of pose cutting
    Fig. 6. Diagram of pose cutting
    Experimental facility
    Fig. 7. Experimental facility
    Relocation experiment. (a) Experimental environment; (b) distance image
    Fig. 8. Relocation experiment. (a) Experimental environment; (b) distance image
    Relocation error. (a) Point A; (b) point B; (c) point C; (d) point D
    Fig. 9. Relocation error. (a) Point A; (b) point B; (c) point C; (d) point D
    Location estimation experiment. (a) Robot path; (b) positioning trajectory of different algorithms
    Fig. 10. Location estimation experiment. (a) Robot path; (b) positioning trajectory of different algorithms
    Positioning accuracy error. (a) Proposed method; (b) Cartographer; (c) Gmapping
    Fig. 11. Positioning accuracy error. (a) Proposed method; (b) Cartographer; (c) Gmapping
    DatasetTotal duration of the dataset /sTotal time of Cartographer /sTotal time of proposed method /sEfficiency ratio
    ACES13664122.31.8
    Intel269117937.64.7
    MIT Killian Court767819039.04.9
    Table 1. Comparison of location estimation efficiency
    MethodIntelACESMIT Killian Court
    Translation errorTranslation varianceTranslation errorTranslation varianceTranslation errorTranslation variance
    Proposed method0.01810.00080.03650.00120.05830.0049
    Cartographer0.02990.00110.03750.00320.03950.0039
    Gmapping0.0310.0020.0440.0040.0500.006
    Table 2. Comparison of positioning estimation error
    ParameterValue
    Measuring radius /m0.15‒12
    Sampling frequency /1038
    Scanning frequency /Hz5.5
    Angular resolution /(°)≤1
    Scanned area /(°)360
    Accuracy of ranging

    actual distance 1% (≤3 m)

    actual distance 2% (3‒5 m)

    actual distance 2.5% (5‒12 m)

    Table 3. RPLIDAR A1 lidar parameter
    ParameterAMCLProposed method
    x /my /mθ /radx /my /mθ /rad
    Absolute value of max error0.0680.0450.0260.0230.0260.024
    Absolute value of min error0.0170.0060.0050.0010.0020.002
    Absolute value of mean error0.0450.0280.0130.0100.0120.012
    Table 4. Relocation error comparison
    Location methodMaximum error along x-axis /mAverage error along x-axis /mRMSE along x-axis /mMaximum error along y-axis /mAverage error along y-axis /mRMSE along y-axis /mRunning time /s
    Gmapping0.3850.07860.11470.2070.08190.097511.0
    Cartographer0.1560.07070.08270.2220.07130.089413.3
    Proposed method0.20.02210.03570.1330.03560.04853.7
    Table 5. Comprehensive comparison of positioning errors among three algorithms