Mahindra Rise Prize: Self Driving Car

Started: June 2016

Stopped: Not yet

Areas worked on: Robotics, Control Systems, Computer Vision, Communication

Current Tasks: Road Bumper Detection using Stereo, Stereo Visual Odometry, Wheel Encoder Fabrication

Team AGV is one of the 13 teams selected for the finals of the Mahindra Rise Prize Challenge. For the same purpose, an electric vehicle namely a Mahindra E2O has been provided to our research group, pending automation. The challenge requires a lot of problem statements to be solved using the car, including a few in their nascent stage in the world of SDCs.

e2o

Lane driving, recognizing traffic signs, traffic lights, creep-crawl to traverse road bumpers are a few challenges for the first round of the competition.

All of these algorithms are first being tested on a dune buggy platform which we have modified to control via a laptop, using a servomotor based throttle, a heavy duty motor based steering and another one for braking, controlled via a Roboteq controller with ROS integration.

I am working in the vision and control team, to solve a few of these problem statements. I have recently implemented road bumper detection using monocular cameras, and am now planning to port the work into stereo vision based detection.thirdres

I also am working on developing a control strategy implementing the drive by wire system already on the car, and allowing for more precise low level control of the vehicle using contactless external encoders. (They are required as the internal feedback on the car has a least count of 1 kmph).

CAN communication is another thing that is being handled by me. We have written communication nodes for the same.

Due to the challenge constraints, I sadly can’t share entire codes on my blog but I’d still post a few relevant posts every now and then.

This entry was posted in Uncategorized. Bookmark the permalink.