Transcript
Page 1: Following Autonomous Mobile Robot

Following Autonomous Mobile Robot

Page 2: Following Autonomous Mobile Robot

Outline

機器人架構 跟隨機器人運作流程 LED 法 ODCS-SIFT 法 (hybird 法 )

Page 3: Following Autonomous Mobile Robot

機器人架構

Page 4: Following Autonomous Mobile Robot

機器人架構 連線方法

Page 5: Following Autonomous Mobile Robot

跟隨機器人運作流程

Page 6: Following Autonomous Mobile Robot

Human Following Behavior of an Autonomous Mobile Robot Using Light-Emitting Device

Page 7: Following Autonomous Mobile Robot

LED 法

Page 8: Following Autonomous Mobile Robot
Page 9: Following Autonomous Mobile Robot
Page 10: Following Autonomous Mobile Robot

參數說明 L ︰兩個 LED 在影像上的距離 u ︰兩個 LED 在影像上與縱軸的距離 d ︰ LED 與 robot 實際的距離 θ ︰ the angle made X straight line

connecting the camera and the device.

Page 11: Following Autonomous Mobile Robot

參數間的關係

Lad /cbu

Page 12: Following Autonomous Mobile Robot

LED 法改進 1.wide-angle lens or omni-directional

camera 2.using more cameras pointing to diff

erent directions. 3.making the camera turning on.

Page 13: Following Autonomous Mobile Robot

LED 法流程圖

Page 14: Following Autonomous Mobile Robot

ODCS-SIFT 法 Object Dominant Color Set (ODCS):it c

an indicate the original image with least color feature.

Scale-invariant feature transform (or SIFT) :an algorithm in computer vision to detect and describe local features in images. The algorithm was published by David Lowe in 1999.

Page 15: Following Autonomous Mobile Robot

ODCS 與 SIFT 範例

Page 16: Following Autonomous Mobile Robot

SIFT 範例 (1/6)

Page 17: Following Autonomous Mobile Robot

SIFT 範例 (2/6)

Page 18: Following Autonomous Mobile Robot

SIFT 範例 (3/6)

Page 19: Following Autonomous Mobile Robot
Page 20: Following Autonomous Mobile Robot

SIFT 範例 (5/6)

Page 21: Following Autonomous Mobile Robot
Page 22: Following Autonomous Mobile Robot

Verification voting

φ= 某顏色出現的比例

ψ= 原圖與 ODCS-database 中的 圖 , 做每一個像素的 ”bitwise AND” 比較後 , 相同的 比例 .

Page 23: Following Autonomous Mobile Robot
Page 24: Following Autonomous Mobile Robot

ODCS-SIFT 流程

Page 25: Following Autonomous Mobile Robot

Experiment

Page 26: Following Autonomous Mobile Robot

Conclusion

This is a foundation for robot to detect object.

The main shortage is all of the parameters are off-line calculated and cannot update automatically.


Top Related