Butterfly Optimization Algorithm(s)

Implementation of the Butterfly Optimization Algorithm and some of its variants.

1. BOA: The standard BOA algorithm from [Arora et al. 2019] & [Arora et al. 2016]
2. mBOA: BOA with intensive search from [Arora et al. 2018]
3. ABOA: BOA with a non-linear update rule for the sensory modality from [Zhang et al. 2020]
4. SABOA: BOA with Self-Adaptative update rule from [Fan et al. 2020]
5. xBOA: BOA with crossover operator from [Bendahmane et al. 2022]

Dependencies

Please install the following packages to use this repository

pip install numpy
pip install pyyaml
pip install pygmo

How to use

1. Change the parameters in the file params.yaml

   • A description of each parameter is given as a comment in the .yaml file
   • You can remove a parameter by commenting or removing its line, it will be replaced by the default value automatically

2. Launch the main script

   python main.py
   

3. You will get an output similar to this:

   Solving problem using xBOA: Crossover Butterfly Optimization Algorithm (UDA)
     Gen    Fevals     Fbest       Improv     Duration
      1       11    900.098353  -10.484016      0.0
      2       16    585.152347  -314.946006     0.0
      3       21    331.047822  -254.104524     0.0
      4       26    177.079935  -153.967888     0.0
      5       32    147.851095   -29.228839     0.0
      6       39    147.851095     0.0          0.0
      7       44     71.580096   -76.270999     0.0
      8       49     42.675169   -28.904927     0.0
      9       55     35.876333    -6.798837     0.0
      10      61     19.645902   -16.23043      0.0
   Executed: 10 generations
   Best solution: [ 0.12479103  0.08723433  0.          0.44437604  0.17309622  0. 0.25159349  0.32912562  3.50034695  0.          0.         11.41293089 1.46827132  0.          0.          0.29118183  0.          1.56295445 0.          0. ]
   

Experimental results and data

We used these algorithms to solve various problems and evaluated their performances. The main contributions are presented in this section:

1. Unknown Area Exploration problem

In [Bendahmane et al. 2022], we used the Butterfly Optimization Algorithm (BOA) and its variants to solve the robotics Unknown Area Exploration problem with energy constraints in dynamic environments. As far as we know, BOA was never used for solving robotics problems before this work.

We also introduced the xBOA variant for the first time, and conducted several experiments for comparing its performance with BOA and the other variants, as well as 5 other metaheuristics used in the literature (GA, PSO, ABC, GWO, CMA-ES).

We found that BOA can be a good alternative to many metaheuristics in terms of the exploration time, while xBOA is more robust to local optima; has better fitness convergence; and achieves better exploration rates than the original BOA and its other variants.

Experimental data and screenshots are available in the folder data/robot-exploration. The full simulation code will be available soon.

2. Path Planning

...to be completed.

3. Numerical Optimization

...to be completed.

Contribute

Contributions are welcome. You may help with the following:

• Writing documentation & tutorials
• Adding new algorithms
• Improving code & performances
• Writing automated tests
• Debugging
• Translation

Please open a new issue for each contribution before sending a pull request.

References

• [Arora et al. 2016]:

  Arora S, Singh S (2016) "An improved butterfly optimization algorithm for global optimization". Advanced Science, Engineering and Medicine 8(9):711–717

• [Arora et al. 2018]:

  Arora S, Singh S, Yetilmezsoy K (2018) "A modified butterfly optimization algorithm for mechanical design optimization problems". Journal of the Brazilian Society of Mechanical Sciences and Engineering 40(1):1–17

• [Arora et al. 2019]:

  Arora S, Singh S (2019) "Butterfly optimization algorithm: a novel approach for global optimization". Soft Computing 23(3):715–734

• [Bendahmane et al. 2022]:

  Bendahmane A, Tlemsani R (2022): "Unknown area exploration for robots with energy constraints using a modified Butterfly Optimization Algorithm". Soft Computing https://doi.org/10.1007/s00500-022-07530-w

• [Fan et al. 2020]:

  Fan Y, Shao J, Sun G, et al (2020) "A self-adaption butterfly optimization algorithm for numerical optimization problems". IEEE Access 8:88,026–88,041

• [Zhang et al. 2020]:

  Zhang M, Long D, Qin T, et al (2020) "A chaotic hybrid butterfly optimization algorithm with particle swarm optimization for highdimensional optimization problems". Symmetry 12(11):1800