Kei18/toio-tswap

toio-tswap

Online TSWAP implementation using the toio robots. It is mainly written in Node.js with yarn build and uses C++(17) with CMake (v≥v3.16) build for the target assignment. The implementation also includes MCPs [1]. They are tested on MacOS 10.15.

Demo

Install

git clone --recursive https://github.com/Kei18/toio-tswap.git
cd toio-tswap
yarn install
yarn build
cmake -B build ./target_assignment/
make -C build

Usage

Step1: Switch on your toio robots and Bluetooth of your PC
Step2: Get the id of the toio robots

yarn run get_id 3

You will get the result, e.g.,

xxxxxxxx 0 green
yyyyyyyy 1 blue
zzzzzzzz 2 red

Step3: Edit the id in the problem file ./sample/problem.yaml
Step4: You may also need to adjust coordinates, see ./sample/graph.yaml
Step5: Set the toio robots in appropriate positions
Step6: Execute!

yarn run tswap ./sample/graph.yaml ./sample/problem.yaml

You can also run MCPs. Do not forget to edit the id in ./sample/plan.yaml:

yarn run mcps ./sample/graph.yaml ./sample/plan.yaml

Experimental Setting

The initial demo is in sample/. For the others, see exp_settings/. I make the id of toio nonpublic, so you have to change the id when you test the settings. Note that the demo with delays is not included but it is easy to do so by modifying src/common.js.

Licence

This software is released under the MIT License, see LICENCE.txt.

Notes

This repository is forked from toio.js. To make the repo private temporarily, I duplicated the original repo.
The target assignment algorithm uses the Dijkstra algorithm instead of BFS. Therefore, the code is not limited to the grid environment, rather, it can work on graphs with non-uniform edge costs.
To convert data from the tswap repo, see notebooks/. I prepared example codes written in Python (Jupyter Notebook).

Author

Keisuke Okumura is a Ph.D. student at the Tokyo Institute of Technology, interested in controlling multiple moving agents.

Reference

Ma, H., Kumar, T. K., & Koenig, S. (2017). Multi-agent path finding with delay probabilities. Proc. AAAI Conf. on Artificial Intelligence (AAAI)

Kei18 / toio-tswap