Constrained Motion Planning Networks X

Ahmed H. Qureshi, Jiangeng Dong, Asfiya Baig, and Michael C. Yip

University of California San Diego

Resources

Please check out our code release!

https://github.com/JiangengDong/CoMPNetX


  1. OMPL's Implementation of CoMPNetX

  2. Full training and testing datasets

  3. Pretrained models for quick evaluations

Kitchen Environment

In this setting the robot is required to organize the kitchen by i) placing the cans and bottle to the trash; ii) open the cabinet and carefully moving (without tilting) the mugs from inside to the tray; and iii) carefully move (without titling) the pitcher into the cabinet.

The following are the total mean computations time taken by each algorithm for solving all the constrained manipulation tasks:

  • CoMPNetX-RRTConnect:

    • Atlas: 36.47s | TB: 53.66s | Projection: 45.57s

  • RRTConnect:

    • Atlas: 194.17s | TB: 217.07s | Projection: 235.26s

Example 1


Example 2


Example 3


Bartender Environment

In this setting the robot is required to clean the tables by placing the cans and bottle to the trash, and carefully moving (without tilting) the mug and kettle to the tray.

The following are the total mean computations time taken by each algorithm for solving all the constrained manipulation tasks:

  • CoMPNetX-RRTConnect:

    • Atlas: 15.06s | TB: 21.35s | Projection: 15.94s

  • RRTConnect:

    • Atlas: 33.44s | TB: 50.01s | Projection: 39.45s

Scenario 1: Example 1


Scenario 1: Example 2


Scenario 1: Example 3


Scenario 2: Example 1

Scenario 2: Example 2

Scenario 3: Example 1

Scenario 3: Example 2

CoMPNetX Sampling


Traditional Uniform Sampling


Bibliography

@article{qureshi2021constrained,

title={Constrained Motion Planning Networks X},

author={Qureshi, Ahmed H and Dong, Jiangeng and Baig, Asfiya and Yip, Michael C},

journal={IEEE Transactions on Robotics},

year={2021}

}

@article{qureshi2020compnet,

author={A. H. {Qureshi} and J. {Dong} and A. {Choe} and M. C. {Yip}},

journal={IEEE Robotics and Automation Letters},

title={Neural Manipulation Planning on Constraint Manifolds},

year={2020},

volume={5},

number={4},

pages={6089-6096}

}