7
16-735 Project Progress Presentation Coffee delivery mission Nov, 13, 2007 NSH 3211 Hyun Soo Park, Iacopo Gentilini 1

16-735 Project Progress Presentation Coffee delivery mission

Embed Size (px)

DESCRIPTION

16-735 Project Progress Presentation Coffee delivery mission. Nov, 13, 2007 NSH 3211 Hyun Soo Park, Iacopo Gentilini. Landmark detection. Changed the robot design to get better resolution (camera closer to the mirror): Used three landmark detection in “good regions”.  2. L 1. L 3.  3. - PowerPoint PPT Presentation

Citation preview

Page 1: 16-735 Project Progress Presentation Coffee delivery mission

16-735 Project Progress PresentationCoffee delivery mission

Nov, 13, 2007NSH 3211

Hyun Soo Park, Iacopo Gentilini

1

Page 2: 16-735 Project Progress Presentation Coffee delivery mission

Landmark detection

• Changed the robot design to get better resolution (camera closer to the mirror):

• Used three landmark detection in “good regions”

2

L12 L3

L2

13

Page 3: 16-735 Project Progress Presentation Coffee delivery mission

FLAN algorithm

3

a. Extraction of landmark features in the image:a. m_lmk : RGB mean of landmark region;

b. s_lmk : RGB standard deviation;

c. d_lmk : Euclidean distance of a pixel when its RGB is different from m_lmk about 1.96 · s_lmk .

b. Calculating landmark regionsa. d : calculate Euclidean distance d for each pixel to

m_lmk and scaling each component by s_lmk;

b. if d ≤ d_lmk pixel is in the landmark region and update value to 255 else update value to 0.

c. Get landmark position

a. for each pixel = 255 check neigh- borhood: if its dimension is bigger than the threshold r_lan, calculate the center of mass of the neighborhood

o position given by built in sensors

x calculated position using three landmark detection

Page 4: 16-735 Project Progress Presentation Coffee delivery mission

Path planning : D* algorithm

- Nomad Simulator : MATLAB

Page 5: 16-735 Project Progress Presentation Coffee delivery mission

Path planning : D* algorithm

- Program on Nomad according to simulator (C++)

- Problem- Noisy sonar : filter required- Robot size

Page 6: 16-735 Project Progress Presentation Coffee delivery mission

Motion algorithm

a. LOAD map

b. SET D* backpointer

c. REPEAT

1. MOVE to TargetPoint

2. READ sonar;

3. state estimation using Kalman filtering;

4. if robot is in a landmark region, update state vector using landmark information;

5. UpdateMap and D*

6. UpdateTargetPoint

7. IF current position == goal position, THEN return

d. END REPEAT

6

Page 7: 16-735 Project Progress Presentation Coffee delivery mission

Scheduling

- Kalman filtering- position level filtering using sonar data and wheel encorder

- Combining image processing algorithm