yeung8023 / model_predictive_conrol-ltl Goto Github PK

This package contains implementations for online-plan synthesis algorithm give a dynamic environment (as the finite transition system) and a potential infeasible Linear temporal logic formula (as the robot’s task). It outputs the predicted trajectories at each time-step that optimize the reward-related objectives and finally fulfill the task maximumly.

The project applies a standard model predictive control based on the P_MAS_TG implementation provided at P_MAS_TG and RHTLC. The reference of P_MAS_TG and RHTLC implementation:

• Guo, Meng, and Dimos V. Dimarogonas, "Multi-agent plan reconfiguration under local LTL specifications", The International Journal of Robotics Research 34.2 (2015). [PDF]
• Ding, Xuchu, Mircea Lazar, and Calin Belta, "LTL receding horizon control for finite deterministic systems", Automatica 50.2 (2014). [PDF]

Project Webpage

1. In the simlulations, the graph-like environments are tested.

   Demo: Simulation Vidoe

2. In the experiment, the neighbor numbers of simulation snapshot provide the time-varying reward (Here we use the random values) which is our optimization objective.

   Demo: Experiment video

• Cai, Mingyu, Hao Peng, Zhijun Li, Hongbo Gao, and Zhen Kan., "Receding Horizon Control-Based Motion Planning With Partially Infeasible LTL Constraints.", IEEE Control Systems Letters 5, no. 4 (2020): 1279-1284. [PDF]
• Cai, Mingyu, Hao Peng, Zhijun Li, Hongbo Gao, and Zhen Kan., "Receding Horizon Control-Based Motion Planning With Partially Infeasible LTL Constraints.", arXiv:2007.12123 (2020). [PDF]

• Allow both normal and infeasible LTL based product automaton task formulas
• Motion model can be dynamic and initially unknown
• Soft specification is maximumly satisfied.
• Online-Path planning is designed from the model predicted control methodology.
• Collect and transfer the real-time data via Optitrack camera systems
• Allow automatically calibrate the mobile robots to obtain its orientation and dynamics.

• Install python packages like networkx2.0.ply
• Add to your PYTHONPATH, to import it in your own project.
• ltlba_32 and ltlba_64 are executable files complied under Linux, please follow [ltl2ba/README.txt]

Try [path_plan.py](https://github.com/mingyucai/Model-Predictive-Control/blob/master/path_plan.py) 

• Add folder to your PYTHONPATH.

Execute [Receding_Horizon_Control.py](https://github.com/mingyucai/Model-Predictive-Control/blob/master/Matlab_simulation/LTL_MPC/RHC_ACC.m)

Please use this bibtex entry if you want to cite this repository in your publication:

@article{cai2020receding,
  title={Receding Horizon Control-Based Motion Planning With Partially Infeasible LTL Constraints},
  author={Cai, Mingyu and Peng, Hao and Li, Zhijun and Gao, Hongbo and Kan, Zhen},
  journal={IEEE Control Systems Letters},
  volume={5},
  number={4},
  pages={1279--1284},
  year={2020},
  publisher={IEEE}
}