Skip to main content

AI-based Energy Harvesting Navigation

Description​

AI solution for energy harvesting aware path planning in a swarm of robots.

  • Primary Functionality: Energy-aware path planning.
  • Target: Server / PC / Robot
  • Task: T3.4
  • Responsible: Mohmmadsadegh Mokhtari

🔗 OpenSwarm-EU AI-based-Energy-harvesting-Navigation Repository

Overview​

This repository contains the Energy Harvesting Aware Path Planning Python codes. It uses PPO and Q-learning methods to find the best path inside the environment while balancing distance and energy harvesting opportunities.

Figure 1: Visualization of the technical workflow for setting up the environment.

This research leverages OpenAI Gymnasium to simulate the environment and PPO learning. Traditional Q-learning is also adapted to the problem. The simulation analyzes different scenarios, including:

  • Grid size
  • Light sources
  • Harvesting range
  • Multiple robots
  • Reward weight impact on navigation performance

Figure 2: Illustration of the learning process outcomes, highlighting efficient pathfinding while maximizing energy harvesting.

Comprehensive results are documented in deliverable T3.4.

Installation Guide​

Step 1: Install Required Dependencies​

Run the following command to install the necessary libraries:

pip install gymnasium 
pip install numpy
pip install pygame

Step 2: Verify Installation​

To confirm that everything is installed correctly, open Python and run:

import gymnasium as gym
from gymnasium import spaces
import numpy as np
import pygame

If no errors appear, the installation was successful.

Step 3: Running the Code​

Execute the following test scripts to see results:

/AI-based-Energy-harvesting-Navigation-main/tests/gridsize_tests.py    
                                                  multi_robot_tests.py 
                                                  reward_weights_tests.py
                                                  lightsource_tests.py 
                                                  range_EH_tests.py