Videos

Weed control using robotic knife in tomato field at UC Davis Farm. (May, 2018)

The experiment conducted in May, 2018 to control Weed automatically at UC Davis tomato field with dense weed population. A novel crop signaling technique is used to detect the crop using advanced machine vision technique.
Eye-in-hand Image based visual servoing with centroid as features. (October, 2017)

Experiment performed on UR5 manipulato for IBVS with centroid of object as feature. Based on the current and desired position of the feature point control law is gives the joint angle velocity of the robot and accordingly manipulator is moving towards the object by minimizing the error.

Language: C++.

Hardware: UR5 Robot, Kinect camera.

Group Member: Mithun P, Deepak Raina

Eye-in-hand Image based visual servoing with 4-points as features. (October, 2017)

Here, green 4-dots are the desired features position and the current features position is red 4-corner points of blue object. The desired feature corner points of blue object is obtained through Kinect camera mounted on end-effector of the UR5 manipulator. The main objective here is to minimize the error between desired and actual feature points. Using the Image Based Visual Servoing we are solving the problem. To minimize the error robot end-effector is moving in upward direction finally reach the desired position.

Software platform: Gazebo, MoveIt!, ROS, OpenCV, PCL, Ubuntu 14.04.

Language: C++.

Hardware: UR5 Robot, Kinect camera.

Eye-in-hand Image based visual servoing with 4-points as features. (March, 2017)

Here, green 4-dots are the desired features position and the current features position is red 4-corner points of blue object. The desired feature corner points of blue object is obtained through Kinect camera mounted on end-effector of the UR5 manipulator. The main objective here is to minimize the error between desired and actual feature points. Using the Image Based Visual Servoing we are solving the problem. To minimize the error robot end-effector is moving towards the desired feature points and finally reaches there.

Here, green dot is the desired feature position and red dot is current feature position of blue object. The main objective here is to minimize the error between desired and actual feature point. Using the Image Based Visual Servoing we are solving the problem. To minimize the error robot end-effector is moving in upward direction finally reach the desired position.

Picking task using UR5 (November, 2016)

APC 2016 Stow Task Video: IITK-TCS Team held at Germany.

Team Members

IITK Laxmidhar Behera, K. S. Venkatesh, Samrat Dutta, Ravi Prakash, Anima Mazumdar, Tharun Reddy, Ranjit Nair, Ashish Kumar and Komal

TCS Swagat Kumar, Ehtesham Hassan, Rekha Raja, Nishant Kejriwal, Sharath Jotawar, Olyvia Kundu and Rajesh Sinha

Result: Hold 5th position out of 16 team.

Picking task using XWAM-Barret Arm at IITK, India (May, 2016)

Stow task for Amazon Picking Challenge (May, 2016)

Experiment with All terrain rover on rough/uneven terrain at Mechatronics Laboratory, IIT Kanpur, India ‎(2015)‎

The path planning problem for a mobile robot operating on rough terrains with six degrees of freedom is far more complex than that of a differential-drive robot operating in a flat terrain with three degrees of freedom. The mobile robot operated

on rough terrain raises difficulty, as mechanisms of mobile robot becomes more complex with their more sophisticated capabilities. Furthermore, the wheel-terrain interactions

and the kinematic structures of the mobile robots, operated on rough terrains, is much more complicated than that of the mobile robots operated on flat terrains, so, it increases

the difficulty in path planning. The rover was experimentally evaluated for following the optimal paths as found using the proposed new potential field method for rough terrain.

it shows the side view of the rover while executing the desired path on rough terrain.

MATLAB simulation for trajectory planning of rover manipulator on flat terrain (Part of my PhD work, 2014).

Mobile manipulation – the coordinated use of manipulation capabilities and the platform degrees of mobility makes the system more versatile and efficient with regard to their

individual motion skills.

The motivation is to determine a feasible path of the rover and joint angle trajectory (θ(t), θ(t)) which reproduces the given trajectory from the start to the end while traversing on the rough terrains.We formulate this problem as an optimization problem and an Augmented Lagrangian Method is used to solve a non-linear constrained optimization. A 14-DOF redundant mobile manipulator is considered for the simulation. Therefore while tracking the end-effector trajectory it's redundancy is need to be solved.

MATLAB simulation for trajectory planning of rover manipulator on uneven terrain. (2014)

Experimental evaluation of trajectory planning of rover (2014)

Experimentation of 4 DOF arm made by Rekha Raja

Page updated

Google Sites

Report abuse