- Log in to post comments
Sentinel Guard: Rover + AI for Rapid Wildlife Identification
TRUMAN MIDDLE SCHOOL
Team Members:
Brian Rios
Abraham Montijo
Karin Urbina
Problem Definition
Wildlife rangers have a hard job because they must watch large areas where animals live. Sometimes the land has tall grass, many animals together, or poor visibility. This makes it difficult to identify animals correctly and on time. When rangers miss animals, injured wildlife or poaching activity may not be noticed quickly.
Our team is creating a project called Sentinel Guard to help solve this problem. Sentinel Guard uses a rover and artificial intelligence (AI) to identify animals automatically. This system can help rangers collect better information and protect wildlife more effectively.
Plan for Solving the Problem
Our solution is a rover built using VEX V5 robotics equipment. The rover has four wheels and is controlled by the VEX V5 Brain. It uses distance sensors and optical sensors to avoid obstacles and move safely. A camera is mounted on top of the rover to take pictures of animals.
The rover uses an AI model to identify five animals: elephant, zebra, giraffe, lion, and wildebeest. We are creating a dataset using open-source wildlife images and pictures we collect ourselves. With help from the Supercomputing Challenge, we train a neural network to recognize these animals. After training, we make the model smaller and faster so it can run on the rover.
When the rover sees an animal, it takes a picture and sends it to the AI model. The system identifies the animal and shows how confident it is. If the confidence is high, the rover saves the time and location, records a short video, and sends an alert.
Progress So Far
So far, we have started building the rover using VEX V5 parts, including the wheels and sensors. We have researched how AI works and how it can be used to recognize images. We have also begun collecting animal images for our dataset. In addition, we have planned how the rover will store information and send alerts.
Expected Results
We expect Sentinel Guard to identify animals with at least 75% accuracy. The AI model should work in under 300 milliseconds, which means it will be fast enough for real-time use. We also expect the rover to move safely and not disturb animals. This project should help rangers protect wildlife and reduce poaching.
Ethics and Safety
Our team will make sure the rover keeps a safe distance from animals and does not interfere with them. All data collected will be used only for science and conservation.
References
- World Wildlife Fund. (2023). Wildlife conservation and monitoring.
https://www.worldwildlife.org - National Geographic. (2022). Technology used to protect wildlife.
https://www.nationalgeographic.com - IBM. (2023). What is artificial intelligence?
https://www.ibm.com/topics/artificial-intelligence - VEX Robotics. (2024). VEX V5 Robotics System.
https://www.vexrobotics.com - Supercomputing Challenge. (2024). Student project resources.
https://supercomputingchallenge.org