Fig. 1. Software system architecture

Fig. 2. Hardware system architecture

Fig. 3. Flowchart of d-Hash algorithm

Fig. 4. Flowchart of key point extraction

Fig. 5. Schematic of the distribution of key points^[13]

Fig. 6. The distribution of human body key points in the spatial coordinate system

Fig. 7. Key point depth estimation effect display. (a) Raw image, (b) key extraction, (c) depth estimation

Fig. 8. Connecting rod coordinate system

Fig. 9. Kinematic modeling of the Mecanum wheel

Fig. 10. Virtual-reality module node distribution

Fig. 11. Scene 3D display effect

Fig. 12. System response speed test

Fig. 13. Test motion trajectory. (a) Beginning of movement; (b) moment in the middle of movement; (c) end of movement

Fig. 14. Motion trajectory test results

Fig. 15. System operation effect display. (a) Robot initialization; (b) the arm turns to the right and retracts; (c) the arm turns to the left and extends forward; (d) the arm turns to the left and clamps the gripper