This project develops an advanced autonomous navigation system for agricultural robots to identify and remove infected plants. Using a Pixhawk microcontroller, Swift Navigation RTK GPS, and 3D lidar, the AGV navigates precisely to infected plants based on UAV coordinates. In the next phase, AGVs and aerial drones will be equipped with advanced sensors like 4D lidar, ultrasonic sensors, depth cameras, RGB cameras, thermal cameras, radar, and RTK GPS to optimize mapping and navigation. This system enhances AGV performance through improved sensor fusion and real-time data acquisition, advancing precision agriculture robotics.

Our team has developed a novel automatic roguing mechanism for AGVs, aimed at minimizing pesticide use and reducing chemical impacts on agricultural products and the environment. Equipped with dual cameras for detecting and removing infected plants, this mechanism effectively reduces crop loss and prevents further infection. A 1/3rd scale prototype was successfully tested in a lab setting. This innovation represents significant advancements in agricultural automation and plant disease management. 

As part of this research, an innovative chassis for autonomous ground vehicles has been developed, optimized for rough terrain navigation. Utilizing Finite Element Methods and MATLAB, the design achieves a lightweight structure with high strength by incorporating nature-inspired patterns and materials like Nylon carbon fiber 12. This chassis outperforms traditional designs, highlighting the potential of additive manufacturing to create robust, lightweight structures and advance autonomous vehicle technology for specialized environments.

Our research team developed an EMG-controlled wire feeder mechanism for hand amputees in TIG welding, addressing skilled trade workers' unique needs. Utilizing electromyography for precise muscle control and exploring force myography for added robustness, this solution enhances prosthetic usability and productivity. Next, we plan to create adaptive prosthetic hands integrating advanced sensor fusion systems with EMG, FMG, MMG, and NIRS signals, leveraging augmented reality for training and calibration. This approach aims to provide enhanced dexterity, adaptability, and user satisfaction, significantly advancing prosthetic technologies for labor-intensive industries.

The GO2 EDU Plus dog robot is a versatile mobile platform. This robot, weighing 15 kg, has a 10 kg payload and a battery life of 2-4 hours. It supports various sensors, including 3D LIDAR, depth, and thermal cameras, and can reach speeds of up to 5 km/h while navigating challenging terrains like stairs. Its advanced capabilities in navigation, mapping, and object detection ensure high reliability and autonomy. 

The project leverages 3D XT16 LIDAR and Astra depth cameras for Simultaneous Localization and Mapping (SLAM) within a ROS framework. By integrating these advanced sensors, the system achieves high accuracy in navigating and mapping its surroundings.

The system integrates high-precision thermal and RGB cameras with machine learning algorithms, employing YOLOv8 for advanced real-time object detection and classification using combined RGB and thermal imagery. This setup enables the robot to distinguish between individuals, significantly improving security through intelligent and autonomous surveillance. 

This research presents the design and implementation of a flexible robotic gripper for articulated robotic arms, addressing the limitations of conventional grippers. The novel compressible gripper features a 14 cm × 6 cm grasping area, capable of handling objects with a width difference of 7 cm and a maximum thickness of 15 cm. Utilizing 3D-printed thermoplastic polyurethane (TPU), the gripper's flexible parts can be interchanged for versatility. It operates without feedback control, enhancing user-friendliness. The experimental evaluation demonstrated exceptional performance in grasping diverse objects. The scalable gripper integrates seamlessly with various robotic arms, enhancing versatility and efficiency in robotic grasping systems. 

This project involves an innovative rotary pin-array gripper mechanism designed to grasp objects of varying sizes and shapes without requiring feedback control. The cylindrical gripper, powered by a stepper motor with a gearbox for enhanced torque, utilizes three stacked curvilinear and linear rails to convert rotational motion into linear motion. The grasping component, comprising three curved parts with numerous compression springs, effectively handles objects between five to nine centimeters in diameter and up to ten centimeters in height. Based on a detailed CAD model and subjected to motion, topology, and stress analyses, a functional prototype has been successfully constructed and tested, demonstrating its capability within the specified size range.

Our team designed and implemented an automatic tool-changing mechanism for industrial and collaborative robotic arms, featuring a universal end-effector connector, diverse tools, and a customized toolbox for autonomous tool switching. The universal connector, equipped with a mechanical lock mechanism and electrical and pneumatic inlets, mounts on the robotic arm, while the toolbox organizes tools within the robot's workspace. Future enhancements will focus on optimizing tool design for reduced size and weight, integrating additional sensors for improved object detection and obstacle avoidance, and enabling tools to receive power and control through the robotic arm. This research advances fully autonomous industrial robotic systems through advanced kinematic principles and optimized mechanical design.

In our lab, students gain hands-on training in programming and operating industrial and collaborative robots. Our team is developing an Augmented Reality-Assisted Robotic System for both industrial tasks and surgical guidance. This system overlays real-time data and 3D imaging onto the physical workspace using AR, enhancing precision and efficiency. In industry, it streamlines operations, reduces errors, and improves safety. In surgical settings, it provides surgeons with enhanced visualization and haptic feedback, leading to more accurate and minimally invasive procedures. This project bridges industrial automation and medical robotics, providing significant advancements in both fields.