Projects
Aerial robots (AR) can be used for surveys, logistics, inspection, etc. However, with current battery technologies, they have very limited operation times. To overcome this issue, a mobile ground robot (MGR) can used in coordination with the AR to replace its battery, while the AR can help navigation of the MGR by providing aerial maps of the area. This project can be divided into 4 main categories:
Development of planning algorithms for coordinated navigation of the robots
Design and...
The utilization of robotics in agriculture stands as a prominent topic within engineering research. The goal of this project is to investigate the use of robotics, machine vision and learning in agricultural applications.
For starters, we have created an indoor robotic farming setting – motivated by the RoboFarm project.
Currently, there are two ongoing projects within this setting:
· Plant growth monitoring: The knowledge gained from this monitoring will serve as feedback for...
In recent years, thanks to the advancement in the 3D sensing technology enables robotics to use high precision depth information without affecting illumination. In fact, for mobile robots that is crucial because mobile robots need to identify their surroundings to carry out some specific tasks. In our project, we aim to develop a novel 3D object detection and classification method that can work with sparse 3D point cloud data.
The number of service robots increases every year and thus, service robotics has become a popular research topic among researchers. As a core ability, a service robot needs to sense human presence and follow the instructor along the environment. The robot needs to adapt its movement in order not to crash any obstacles or people along the way. While moving, the robot should be able to distinguish different places (e.g rooms in an indoor application) by observing visual properties. Throughout...
The project intends to increase knowledge of robot about its environment. Specifically, we want robot to understand which object a human is pointing with her hand through data taken from Kinect. Both traditional machine vision techniques and deep learning will be incorporated in the project. The job to bo done can be divided to 3 parts:
- Detecting hands of the instructor human
- Checking if hands are in the pointing pose.
- Predicting related object
Minik robots have been designed and developed in ISL. There are 5 Minik robots.
They are differential type and have motion mechanisms.
History:
Fall 2016:
Low level motion control: Updated Arduino-based systems.
Spring 2017:
The first version of vision system is integrated.
Their vision system has been updated.
Their processors are Rasperry Pi 3.
Fall 2018:
These robots are used in EE451 as part of final Project.
EE 451 Final Project Videos
2016-2017:
Group 1
Group 2
Group 3
Group 4
Group 5...
