Author Affiliations
1School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 200444, China2State Key Laboratory of Mechanical Transmissions, Chongqing University, Chongqing 400044, China3Joint Laboratory of High Power Laser and Physics, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, Chinashow less
Fig. 1. The mobile robot
Fig. 2. Pose of a mobile robot in the two-dimensional coordinate system
Fig. 3. Images from laser sensor. (a) Original local image from laser sensor; (b) local image from laser sensor after normalization
Fig. 4. Images after threshold segmentation
Fig. 5. Local maps of different scenes. (a) End of the corridor; (b) straight corridor; (c) corridor corner
Fig. 6. Vertical projection and horizontal projection of local maps in different scenes. (a) End of the corridor; (b) straight corridor; (c) corridor corner
Fig. 7. The robot enters the T-shaped corner in different orientations
Fig. 8. Structure diagram of classifiers
Fig. 9. Straight scene. (a) Straight corridor; (b) straight T-shaped corner
Fig. 10. Interference line segment filtering
Fig. 11. Right turn corner
Fig. 12. End of the corrido
Fig. 13. The experimental environment
Fig. 14. Path planning for straight area
Fig. 15. Path planning for corner area
Fig. 16. Path planning for the end area
Fig. 17. Construction effect of global map. (a) Viewable navigation path; (b) environmental raster map
Classifier | NTP | NFP | NFN | NTN | R /% | P /% |
---|
H1 | 14 | 0 | 1 | 70 | 93.33 | 100 | H2 | 29 | 1 | 1 | 39 | 96.67 | 96.67 | H3 | 10 | 0 | 0 | 30 | 100 | 100 |
|
Table 1. Effects of the classifiers