Replicating Results

[bmrocamora]

Hi, I'm trying to replicate the experiments with DWA and EBand at 0.5m/s, but if I try running run.py with the launchers provided, I see the odometry degrading really fast. The rover is very likely to cross the region with obstacles, but by the time it gets to the open area, it does not converge to 1m of the goal position. Did you see this behaviour?

[Daffan]

I didn't see this before. The odometry reading makes pretty much sense to me. Did you try to run it out of the container with the GUI to see where did the robot go?

[ahanjaya]

I faced the same problem using ubuntu 20 with ros noetic, but the problem didn't occur in ubuntu 18 with ros melodic

[2892510130]

I found the same problem in both Ubuntu 18 and 20. When I run the run.py, there is some deviation between the base_link axis and the odom axis, which causes that the target point in the odom coordinate system is not directly in front of the robot in word coordinates, resulting in navigation errors.

[Su-study]

I found the same problem in both Ubuntu 18 and 20. When I run the run.py, there is some deviation between the base_link axis and the odom axis, which causes that the target point in the odom coordinate system is not directly in front of the robot in word coordinates, resulting in navigation errors.

Unfortunately, I also encountered the same problem. Did you finally solve it?