Robotics Projects

Problem Statement
To reduce manual labor & impact of herbicides by prototyping an agricultural robot for detection & removal of weeds from crops.

Target Audience
Farmers

Overview
• Object Detection using YOLO (You Only Look Once) v5 model
• 550+ images for model training
• iPhone camera input utilized for processing on Macbook laptop*
• Probabilistic exploration mode
• 95% weed removal accuracy
• Hardware implemented using LEGO Mindstorm EV3 kit

* MacOS and iOS based system essential for continuity camera feature by Apple

Tools & Frameworks
Python • PyTorch • OpenCV

• Live Demonstration
• Project Presentation
• Code

Problem Statement
To create a efficient system for exploring and mapping the search field using multiple robots and find the object(s) of interest

Target Audience
Disaster Management (Search & Rescue), Defence (Exploration, Retrieval & Surveillance)

Overview
• Global path planning with combination of Levy walk & Brownian motion
• Local path planning with RRT*
• Motion planning with Non-linear Model Predictive Controller (NMPC)
• Obstacle avoidance using potential gradient obstacle avoidance algorithm
• Generalized local map coverage sharing among robots with ellipse fitting
• Scaleable system (tested for up to 10 robots)
• 85+ simulations over 250m² area for 1-6 robots
• 100% Object finding accuracy with modified algorithms
• Coded in Python

Example
Input: Object randomly placed, select number of robots
Output: See images for reference

Tools & Frameworks
Python • Matplotlib • SciPy • PyLab • NumPy • GitHub

• Code
• Simulation Video

Problem Statement
To simulate a 4 DOF arm and end effector using Modelling & Control of Robotics concepts

Target Audience
Future students (visual learning with easy simulation)

Overview
• Path planning & obstacle avoidance with RRT*
• Trajectory tracking & correction with PI controller
• Inverse Kinematics for joint angles
• Spline Interpolation for motion smoothening
• 95% Accurate modeling of moving the end effector to the desired pose
• 0.1 norm of error
• Coded in MATLAB
• Interface designed in App Designer (MATLAB)

Example
Input: Specify x, y, z coordinates ~ desired pose
Output: See images for reference

Tools & Frameworks
MATLAB • App Designer

• Project Presentation