SauRoN (Simulation for autonomous robot navigation) is a simulation environment for reinforcement learning algorithms, which was developed within the research project ai arena. It is intended to help students to better understand and apply these methods during their studies.
The simulation is currently being further developed as part of a software project.
Agents in the simulation should learn to avoid obstacles and navigate towards their specific goals. They are using laserscans, distance to goal, velocity (linear and angular) and orientation as inputs for the neural network and the network outputs the new velocity commands.
Clone the repository:
git clone [email protected]:RoblabWh/SauRoN.git
Install Anaconda on your system:
Run the install_conda_env.sh script to create your anaconda environment with it. Now you should be ready to run your simulation.
There are some levels provided as svg Files. Red dots mark possible agent positions while green dots mark possible goals for the agents.
Currently only PPO is implemented
The training can be started either using MPI for Multi Environment or without MPI to only train on a single level.
From your conda-environment run main.py with desired arguments:
--modechoose eithertrainortestfor Training or running a trained model--ckpt_folderCheckpoint Folder for the loaded and new models. Default :./models
There are several other options for the hyperparameters, vizualization and robot settings you can look up in the main.py.
There is a trained network provided under ./models. If the installation was successful the following command should run a pretrained network:
python main.py --ckpt_folder ./pretrained --model batch_8 --mode test --level_files 'kreuzung.svg'
To train run:
python main.py --mode train
To train with MPI:
mpirun -n N_PROC python main.py --mode train
--restore: If set to True, restores and continues training from previous checkpoint. Default: False
--time_frames: The number of timeframes (past states) that will be analyzed by the neural net. Default: 4
--steps: The number of steps in the environment per episode. Default: 2500
--max_episodes: The maximum number of episodes to run. Default: inf
--update_experience: How many experiences to use to update the policy. Default: 20000
--batches: The number of batches to use. Default: 2
--action_std: The constant standard deviation for the action distribution (Multivariate Normal). Default: 0.5
--K_epochs: The number of times to update the policy. Default: 7
--eps_clip: The epsilon value for p/q clipping. Default: 0.2
--gamma: The discount factor. Default: 0.99
--lr: The learning rate. Default: 0.0003
--input_style: Choose between using images (image) or laser readings (laser). Default: laser
--image_size: The size of the image that is input to the neural net. Default: 256
--level_files: A list of level files as strings. Default: ['svg3_tareq2.svg']
--sim_time_step: The time between steps. Default: 0.1
--number_of_rays: The number of rays emitted by the laser. Default: 1081
--field_of_view: The lidar's field of view in degrees. Default: 270
--collide_other_targets: Determines whether the robot collides with targets of other robots (or passes through them). Default: False
--manually: Move the robot manually with wasd. Default: False
--visualization: Choose the visualization mode: none, single or all. Default: "single"
--visualization_paused: Start the visualization toggled to paused. Default: False
--tensorboard: Use tensorboard. Default: True
--print_interval: How many episodes to print the results out. Default: 1
--solved_percentage: Stop training if objective is reached to this percentage. Default: 0.99
--log_interval: How many episodes to log into tensorboard. Also regulates how solved percentage is calculated. Default: 30
--render: Whether to render the environment. Default: False
--scale_factor: The scale factor for the environment. Default: 55
--display_normals: Determines whether the normals of a wall are shown in the map. Default: True