UAV Search & Rescue Simulation

Devlog

• NEW: UAV Search & Rescue Simulation Progress
  August 18, 2025 by Shivani

  In this devlog, I showcase my work in progress. I have made a simulation where players can manage drones to issue commands such as "take off" and "take over". This allows the player to have control ov... Continue reading
• NEW: Drone Learns To Complete Search & Rescue Missions
  August 17, 2025 by Shivani

  Environment For training the agent, I used curriculum learning, which allows me to increase the difficulty gradually over training steps. For the curriculum, I decided to create three maps of differin... Continue reading

  
• Devlog #8 - Evaluating Trained Models
  July 21, 2025 by Shivani

  Model Comparison Parameter Default Model SearchAgent (Custom) Description trainer_type ppo ppo Same algorithm used max_steps 500,000 3,000,000 Extended training for more learning and convergence summa... Continue reading
• Devlog #7 - Training My Drone: Curriculum Learning for Search Tasks
  July 15, 2025 by Shivani

  Check out my training video which shows the process of training a drone using curriculum. Training a drone in a complex environment isn't easy, especially when you introduce mechanics like batteries,... Continue reading

  
• Devlog #6 - Training My Drone: Custom PPO Configuration
  July 15, 2025 by Shivani

  Before I started building a custom PPO configuration, I ran into a wall. I had just expanded the drone's environment, adding a battery system, a charging station, and increasing the environment size... Continue reading

  
• Devlog #5 - Training My Drone: Battery Awareness
  July 08, 2025 by Shivani

  After completing the initial version of the drone agent and successfully training it to collect a target, I wanted to push the simulation further. The goal this time was to increase complexity and mak... Continue reading

  
• Devlog #4 - Training My Drone: Target Collection Basics
  June 26, 2025 by Shivani

  The Task The goal of the agent is to navigate toward a target. If the agent successfully reached the target, it received a positive reward of +5. However, if it collided with the boundaries of the env... Continue reading

  
• Devlog #3 - Training My Drone: ML-Agents Setup
  June 26, 2025 by Shivani

  When designing a Multi-Agent Reinforcement Learning (MARL) environment, the choice of simulation platform significantly impacts development speed. Unity provides an ML-Agents toolkit built for reinfor... Continue reading

  
• Devlog #2 - Training My Drone: Flight Controls
  June 26, 2025 by Shivani

  drone controls Scripts public class DroneInputHandler : MonoBehaviour // Properties public Vector2 Cyclic { get => cyclic; } public float Pedals { get => pedals; } public float Throttle { get => throt... Continue reading

  
• Devlog #1 - Final Project Concept
  June 26, 2025 by Shivani

  This project utilises parcel delivery using MARL, where autonomous drone agents learn to collaborate efficiently to complete search and delivery tasks. Human supervisors will be able to issue high-lev... Continue reading