Kinematic Parameter Identification of Industrial Robot Based on Binocular Vision

Yanqiong Shi; Kefan Li; Rongsheng Lu; Xiyong Zhou

doi:10.3788/LOP222047

Journals >Laser & Optoelectronics Progress >Volume 60 >Issue 14 >Page 1415002 > Article

Laser & Optoelectronics Progress
Vol. 60, Issue 14, 1415002 (2023)

Kinematic Parameter Identification of Industrial Robot Based on Binocular Vision

Yanqiong Shi^1、*, Kefan Li¹, Rongsheng Lu², and Xiyong Zhou¹

Author Affiliations

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

²School of Instrument Science and Opto-Electronics Engineering, Hefei University of Technology, Hefei 230009, Anhui, China

show less

DOI: 10.3788/LOP222047 Cite this Article Set citation alerts

Yanqiong Shi, Kefan Li, Rongsheng Lu, Xiyong Zhou. Kinematic Parameter Identification of Industrial Robot Based on Binocular Vision[J]. Laser & Optoelectronics Progress, 2023, 60(14): 1415002 Copy Citation Text

show less

Fig. 1. 3D coordinate system relationship diagram of robotic arm

Download full size

Fig. 2. Binocular vision recognition robot kinematic parameter system

Download full size

Fig. 3. Pseudo-random sequence and Halton sequence comparison chart

Download full size

Fig. 4. Flow chart of kinematic parameter identification by STAPSO

Download full size

Fig. 5. Layout of the experimental device

Download full size

Fig. 6. Measurement error diagrams of binocular vision system. (a) 70 mm error point set; (b) x, y axis 10 mm, 70 mm average error; (c) x-axis range error; (d) y-axis range error

Download full size

Fig. 7. Spatial distribution of collection points

Download full size

Fig. 8. Comparison chart of PSO algorithm iteration

Download full size

Fig. 9. Comparison of identification results of each algorithm

Download full size

j	$α_{i - 1}$ /（ $°$ ）	$a_{i - 1}$ /mm	$θ_{i}$ /（ $°$ ）	$d_{i}$ /mm
1	0	0	$θ_{1}$	144
2	90	0	$θ_{2}$ -90	0
3	0	-264	$θ_{3}$	0
4	0	-236	$θ_{4}$ -90	106
5	90	0	$θ_{5}$	114
6	-90	0	$θ_{6}$	67

Table 1. MDH model parameter table

j	$α_{i - 1}$ /（ $°$ ）	$a_{i - 1}$ /mm	$θ_{i}$ /（ $°$ ）	$d_{i}$ /mm
1	0-0.1746	0	$θ_{1}$ -0.3136	144
2	90+0.6159	0+0.9999	$θ_{2}$ -90-0.9993	0
3	0-0.7798	-264+0.9999	$θ_{3}$ +0.9999	0
4	0-0.5255	-236-0.2770	$θ_{4}$ -90-0.6520	106+0.5261
5	90+0.9999	0+0.2263	$θ_{5}$ +0.1365	114+0.2078
6	-90+0.6489	0	$θ_{6}$	67+0.1015

Table 2. Parameter table of the identified MDH model

Parameter	Before calibration	trust-region（increased accuracy /%）	PSO（increased accuracy /%）	STAPSO（increased accuracy /%）
Maximum error /mm	2.7903	0.8484（69.59）	0.843（69.79）	0.573（79.46）
Minimum error /mm	0.1265	0.055（56.52）	0.051（59.68）	0.0501（60.40）
Average error /mm	1.1601	0.3023（73.94）	0.3164（72.73）	0.226（80.52）
Standard deviation /mm	0.6582	0.2212（66.39）	0.2037（69.05）	0.1412（78.55）

Table 3. Statistical analysis of distance error

Yanqiong Shi, Kefan Li, Rongsheng Lu, Xiyong Zhou. Kinematic Parameter Identification of Industrial Robot Based on Binocular Vision[J]. Laser & Optoelectronics Progress, 2023, 60(14): 1415002

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information