IvLabs

This project aims to control and simulate the turtlebot3 in Gazebo using hand gestures. It has two parts:
Detection of Hand gestures using the OpenCV library.
Using the above code to control the linear and angular velocities of the turtlebot3 in Gazebo.

Detecting the hand gestures:
1. We first create a mask to segment the hand in HSV colour space.
2. Further noise is removed by first applying Morphological transformation on it and then smoothing the image using a Gaussian filter.
3. The contours of the hand are found and then using various properties such as convex hull, convex area and aspect ratio, the various hand gestures are differentiated from each other.

Implementation on turtlebot3:
1. This project uses ROS Noetic.
2. First, a ROS publisher is created in order to control the turtlebot3.
3. This publisher then publishes the Twist message in the rostopic cmd_vel.
4. The above code is used to detect the following hand gestures and corresponding changes are made in the linear and angular velocities of the turtlebot3.
0 - Stationary(velocities set to zero)
1 - Moving forward(increase in linear velocity)
2 - Moving backward(decrease in linear velocity till it becomes negative)
3 - Turning right(increase in angular velocity)
4 - Turning left(decrease in angular velocity till it becomes negative)

Refer the link in our bio for the full code.

Members:
, Deepshikhar Gupta, , , ,

Mentors:
, .singh.09, , Nachiket Sawwalakhe, ,

A journey that started from the stairs of mechanical department in 2012, to our own allocated space in the new academic building in 2019, we've surely come a long way. 😌✨

Even the IvLabs' Facebook account was created around the same time, in 2019. We witnessed the followers count rising from nothing and today here we are, four-figure number of around 1.9k. It's rather a small win, but there’s nothing wrong in celebrating these everyday small achievements, right?😉

🤗 Thank you everyone for joining in and witnessing journey!

This project showcases the Sahayak Robot that was made capable of solving a maze-like environment. The Wall Following algorithm was used to solve a simply connected maze. Sahayak was simulated in a Gazebo environment using the Robot Operating System (ROS).

The team began by applying the Maze Following Algorithm to a single wall by utilizing data from the LiDAR sensor. Samples acquired from the sensor were divided into regions and based on these readings, the maneuvers were decided for the robot. Finally, using the conditions of the Wall Following algorithm and the data collected from the LiDAR sensors, the robot was able to solve the maze environment.

Refer bio for the GitHub link.

Members: Malhar Limaye, , Laeba Sayedthakur, Madhav Ishwar

Mentors: , , .ringe, .subhedar,