MonoForce

Robot-terrain interaction prediction from only RGB images as input.

Table of Contents

• Installation Instructions
• Data
• Terrain Encoder
• Differentiable Physics
• Running
• ROS Integration
• Navigation
• Citation

Running

The MonoForce pipeline consists of the Terrain Encoder and the Differentiable Physics modules. Given input RGB images and cameras calibration the Terrain Encoder predicts robot's supporting terrain. Then the Differentiable Physics module simulates robot trajectory and interaction forces on the predicted terrain for a provided control sequence (linear and angular velocities).

Please run the following command to explore the MonoForce pipeline:

python scripts/run --img-paths IMG1_PATH IMG2_PATH ... IMGN_PATH --cameras CAM1 CAM2 ... CAMN --calibration-path CALIB_PATH

For example if you want to test the model with the provided images from the RobinGas dataset:

python scripts/run --img-paths config/data_sample/tradr/images/1666267171_394104004_camera_front.png \
                               config/data_sample/tradr/images/1666267171_394104004_camera_left.png \
                               config/data_sample/tradr/images/1666267171_394104004_camera_right.png \
                               config/data_sample/tradr/images/1666267171_394104004_camera_rear_left.png \
                               config/data_sample/tradr/images/1666267171_394104004_camera_rear_right.png \
                   --cameras camera_front camera_left camera_right camera_rear_left camera_rear_right \
                   --calibration-path config/data_sample/tradr/calibration/ \
                   --lss_cfg_path config/lss_cfg_tradr.yaml --model_path config/weights/lss/lss_robingas_tradr.pt --dphys_cfg_path config/dphys_cfg.yaml \
                   --linear-vel 0.5 --angular-vel -0.1

Please, refer to the Terrain Encoder documentation to download the pretrained model weights.

If you have ROS and Docker installed you can also run:

docker pull agishrus/monoforce
cd docker/
./run.sh

ROS Integration

We provide a ROS nodes for both the trained Terrain Encoder model and the Differentiable Physics module. They are integrated into the launch file:

roslaunch monoforce monoforce.launch

Navigation

Navigation method only using RGB images. The package is used as robot-terrain interaction and path planning pipeline.

Trajectories prediction is based on the NVIDIA/warp and ctu-vras/diffsim packages.

Navigation consists of the following stages, video:

• Height map prediction: The Terrain Encoder part of the MonoForce is used to estimate terrain properties.
• Trajectories prediction: The Diff Physics part of the MonoForce is used to shoot the robot trajectories.
• Trajectory selection: The trajectory with the smallest cost based on robot-terrain interaction forces is selected.
• Control: The robot is controlled to follow the selected trajectory.

Please, refer to ctu-vras/husky_nav package for the navigation example in Gazebo simulator.

Citation

Consider citing the papers if you find the work relevant to your research:

@article{agishev2023monoforce,
    title={MonoForce: Self-supervised Learning of Physics-aware Model for Predicting Robot-terrain Interaction},
    author={Ruslan Agishev and Karel Zimmermann and Vladimír Kubelka and Martin Pecka and Tomáš Svoboda},
    year={2023},
    eprint={2309.09007},
    archivePrefix={arXiv},
    primaryClass={cs.RO}
}

@inproceedings{agishev2024endtoend,
    title={End-to-end Differentiable Model of Robot-terrain Interactions},
    author={Ruslan Agishev and Vladim{\'\i}r Kubelka and Martin Pecka and Tomas Svoboda and Karel Zimmermann},
    booktitle={ICML 2024 Workshop on Differentiable Almost Everything: Differentiable Relaxations, Algorithms, Operators, and Simulators},
    year={2024},
    url={https://openreview.net/forum?id=XuVysF8Aon}
}