1. CSE-473
Mobile Robot Mapping

2. Mapping with Raw Odometry

3. Graphical Model of Mappingu u u 1
t-1
... x x x x 1 2 t m z z z 1 2 t

4. Graph-SLAM Idea

5. Robot Poses and Scans [Lu and Milios 1997]
Successive robot poses connected by odometry
Sensor readings yield constraints between poses
Constraints represented by Gaussians
Globally optimal estimate

6. Loop Closure Use scan patches to detect loop closure
Add new position constraints
Deform the network based on covariances of matches
Before loop closure
After loop closure

7. Mapping the Allen Center

8. 108 features, 105 poses, only few secs using cg.
3D Outdoor Mapping
108 features, 105 poses, only few secs using cg.

9. Map Before Optimization

10. Map After Optimization

11. Rao-Blackwellized Mapping
Compute a posterior over the map and possible trajectories of the robot :
map and trajectory
measurements
map
robot motion
trajectory

12. … FastSLAM Robot Pose 2 x 2 Kalman Filters Landmark 1 Landmark 2
Landmark N …
x, y, 
Particle #1
Landmark 1
Landmark 2
Landmark N …
x, y, 
Particle #2
Landmark 1
Landmark 2
Landmark N …
x, y, 
Particle #3 …
Particle M
Landmark 1
Landmark 2
Landmark N …
x, y, 
[Begin courtesy of Mike Montemerlo]

13. Example 3 particles map of particle 3 map of particle 1

14. Rao-Blackwellized Mapping with Scan-Matching
Map: Intel Research Lab Seattle
Loop Closure

15. Rao-Blackwellized Mapping with Scan-Matching
Map: Intel Research Lab Seattle
Loop Closure

16. Rao-Blackwellized Mapping with Scan-Matching
Map: Intel Research Lab Seattle

17. Example (Intel Lab) 15 particles
four times faster than real-time P4, 2.8GHz
5cm resolution during scan matching
1cm resolution in final map
joint work with Giorgio Grisetti

18. FastSLAM – Simulation Up to 100,000 landmarks 100 particles
103 times fewer parameters than EKF SLAM
Blue line = true robot path
Red line = estimated robot path
Black dashed line = odometry

19. Victoria Park Results 4 km traverse 100 particles
Uses negative evidence to remove spurious landmarks
Blue path = odometry
Red path = estimated path
[End courtesy of Mike Montemerlo]

20. Frontier Based Exploration
[Yamauchi et al. 96], [Thrun 98]

21. Multi-Robot Mapping
Robot A
Robot B
Robot C