Lidar SLAM Method for Photovoltaic Power Stations Based on Factor Graph Optimization

Fangbin Wang; Kun Cao; Xue Gong; Darong Zhu; Ping Wang

doi:10.3788/LOP232263

Laser & Optoelectronics Progress, Volume. 61, Issue 14, 1415002(2024)

Lidar SLAM Method for Photovoltaic Power Stations Based on Factor Graph Optimization

Fangbin Wang^1,2、*, Kun Cao¹, Xue Gong^1,2, Darong Zhu^1,2, and Ping Wang¹

Author Affiliations

¹School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei 230601, Anhui, China

²Key Laboratory of Construction Machinery Fault Diagnosis and Early Warning Technology of Anhui Jianzhu University, Hefei 230601, Anhui, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(17)

Fig. 1. Cartographer algorithm framework

Download full size

Fig. 2. Schematic of scan matching error

Download full size

Fig. 3. Improved algorithm's framework

Download full size

Fig. 4. IMU pre-integration principle

Download full size

Fig. 5. Factor graph structure

Download full size

Fig. 6. Mobile robot

Download full size

Fig. 7. Simulated photovoltaic power station environment

Download full size

Fig. 8. Comparison of mapping results under simulated photovoltaic power station environment. (a) Gmapping;(b) Cartographer; (c) improved algorithm

Download full size

Fig. 9. Line chart of relative errors for three algorithms in the simulated photovoltaic power station environment

Download full size

Fig. 10. Simulated narrow corridor environment

Download full size

Fig. 11. Comparison of mapping results under simulated photovoltaic power station environment.(a) Gmapping;(b) Cartographer; (c) improved algorithm

Download full size

Fig. 12. Feature point selection in simulated narrow corridor environment

Download full size

Fig. 13. Line chart of relative errors for three algorithms in the simulated narrow corridor environment

Download full size

Table 1. Parameters of LSLIDAR M10
View table
Table 1. Parameters of LSLIDAR M10
Parameter Measurement distance /m Scanning angle /（°） Measurement rate /（point·s^-1） Angular resolution /（°） Scanning frequency /Hz
Value 0.03‒25 0‒360 10000 0.36 10

Table 2. Error analysis of three algorithms in simulated photovoltaic power station environment

View table

Table 2. Error analysis of three algorithms in simulated photovoltaic power station environment

Parameter	Gmapping	Cartographer	Improved algorithm
Max error /m	0.983	0.153	0.037
Min error /m	0.038	0.007	0.008
Mean square error /m²	0.164	0.007	0.0006
Root mean square error /m	0.405	0.086	0.024
Mean absolute error /m	0.254	0.071	0.022
Mean relative error /%	7.538	3.232	0.939

Table 3. Error analysis of three algorithms in simulated narrow corridor environment

View table

Table 3. Error analysis of three algorithms in simulated narrow corridor environment

Parameter	Gmapping	Cartographer	Improved algorithm
Max error /m	3.250	1.956	0.064
Min error /m	0.046	0.010	0.011
Mean square error /m²	1.943	0.024	0.002
Root mean square error /m	1.394	0.155	0.042
Mean absolute error /m	0.966	0.092	0.038
Mean relative error /%	10.735	1.104	0.646

Table 4. CPU load comparison
View table
Table 4. CPU load comparison
Parameter Cartographer Improved algorithm
Mean CPU load /% 30.107 17.765
Median CPU load /% 29.9 17.4
Standard deviation of CPU load /% 4.506 3.482

Tools

Get Citation

Copy Citation Text

Fangbin Wang, Kun Cao, Xue Gong, Darong Zhu, Ping Wang. Lidar SLAM Method for Photovoltaic Power Stations Based on Factor Graph Optimization[J]. Laser & Optoelectronics Progress, 2024, 61(14): 1415002

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites