search

reiniscimurs / DRL-robot-navigation

Deep Reinforcement Learning for mobile robot navigation in ROS Gazebo simulator. Using Twin Delayed Deep Deterministic Policy Gradient (TD3) neural network, a robot learns to navigate to a random goal point in a simulated environment while avoiding obstacles.
MIT License
488 stars 98 forks source link

model [cardboard_box_$$] does not exist #30

Closed mincheulkim closed 1 year ago

mincheulkim commented 1 year ago

Hello, thanks to your great work.

When I compiled the code and run the training, the robot runs on only empty space both RViz and gazebo, not in the kind of static obstacle world.

In the terminal, below error message are shown.

[ERROR] [1662014359.011517132, 2.667000000]: Updating ModelState: model [cardboard_box_0] does not exist [ERROR] [1662014359.011602175, 2.668000000]: Updating ModelState: model [cardboard_box_1] does not exist [ERROR] [1662014359.011676078, 2.668000000]: Updating ModelState: model [cardboard_box_2] does not exist [ERROR] [1662014359.011693132, 2.668000000]: Updating ModelState: model [cardboard_box_3] does not exist

I expected some static obstacles such as wall, but it doesn't shown.

reiniscimurs commented 1 year ago

Hi,

Please check the answer in https://github.com/reiniscimurs/DRL-robot-navigation/issues/16#issue-1215832058

Essentially, check the terminal log for any missing packages and check that the paths to installation have been set up properly. If that does not work, let me know the full log output from the terminal

mincheulkim commented 1 year ago

Hello, I set GAZEBO_RESOURCE_PATH incorrectly. Now the program works fine. thank you!