Tham khảo tài liệu 'contemporary robotics - challenges and solutions part 6', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Simultaneous Localization and Mapping SLAM of a Mobile Robot Based on Fusion of Odometry and Visual Data Using Extended Kalman Filter 141 The name of the lines is related to the temporal order in which the robot identified them. Figure 9 shows the moment of identification of each straight line. Initially the robot identifies two lines Figure 9-a advancing it identifies the third Figure 9-b and a little further on identifies the fourth Figure 9-c . The fifth line is only seen when the robot has already negotiated around half of the trajectory Figure 9-d and the sixth is only identified when the robot begins to rotate on its own axis Figure 9-e . Fig. 9. Temporal process of identifying the floor lines. Considering the total movement of the robot we found that it identifies lines 56 of the time one line 65 of the time and two lines 35 of the time as shown in Figure 10. This figure shows that at a particular instant the robot identified three lines. This incorrect information originated in an image where the shadow of an obstacle was identified as a line. 142 CONTEMPORARY ROBOTICS - Challenges and Solutions Fig. 10. Temporal process of identifying floor lines. The process of matching all the experience was successful in 93 of the cases as illustrated in Table 2. Figure 11 shows the straight lines identified during the robot s trajectory. Based on line 3 the line most seen during the experiment Figure 12 shows the behavior of the p and a variance for this line. It can be seen that the behavior is in accordance with the theory and is always minimized when it and other lines are observed. Simultaneous Localization and Mapping SLAM of a Mobile Robot Based on Fusion of Odometry and Visual Data Using Extended Kalman Filter 143 Fig. 12. Variance behavior a p of line 3 and b a of line 3. The coordinates x y and 0 and their respective variances are shown in Figure 13. a b d c
