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
486 stars 97 forks source link

Observed that the car only performed circular motion #126

Open hhhiiittt opened 3 months ago

hhhiiittt commented 3 months ago

Describe the bug A clear and concise description of what the bug is.

To Reproduce Steps to reproduce the behavior:

Expected behavior A clear and concise description of what you expected to happen.

Screenshots If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

Additional context Add any other context about the problem here.

hhhiiittt commented 3 months ago

Hello, thank you for your work. I trained 110 times and then stopped the training using the terminal command you provided. There are also saved training models in the folder. But when I tested, the car could only perform circular motion without path planning. What is the reason for this? How to solve it? Thank you for your reply! fa50561ac9248555b63779c7393f0e6c ![Uploading 921158d0e00b74394d06b24bbf09fc17.png…]()

hhhiiittt commented 3 months ago

![Uploading 921158d0e00b74394d06b24bbf09fc17.png…]()

reiniscimurs commented 3 months ago

Hi,

See other issues on convergence: https://github.com/reiniscimurs/DRL-robot-navigation/issues/103

hhhiiittt commented 3 months ago

Hi,

See other issues on convergence: #103

thank you

reiniscimurs commented 3 months ago

Btw what do you mean by 110 times? Is it 110 episodes or epochs?

hhhiiittt commented 3 months ago

Btw what do you mean by 110 times? Is it 110 episodes or epochs?

Thank you for your reply.It is 110 episodes

reiniscimurs commented 3 months ago

110 episodes is not enough training time for the model. From my experience you will need 40 to 60 epochs to start seeing any reasonable performance.

hhhiiittt commented 3 months ago

110 episodes is not enough training time for the model. From my experience you will need 40 to 60 epochs to start seeing any reasonable performance.

Understood, thanks for the guidance