Follow us on social

Latest Posts

Stay in Touch With Us

For Advertising, media partnerships, sponsorship, associations, and alliances, please connect to us below


+91 40 230 552 15

540/6, 3rd Floor, Geetanjali Towers,
KPHB-6, Hyderabad 500072

Follow us on social


  /  Latest News   /  Roboticists are Using TartanDrive Data to Make Your Mountain Rides More Adventurous
TartanDrive data

Roboticists are Using TartanDrive Data to Make Your Mountain Rides More Adventurous

Researchers claim that TartanDrive data could be useful for a self-driving vehicle to navigate off-road.

Roboticist researchers from Carnegie Mellon University took an all-terrain vehicle on wild rides through tall grass, loose gravel, and mud to gather data about how the ATV interacted with a challenging, off-road environment. ATVs are motorized off-highway vehicles designed to travel on four low-pressure or non-pneumatic tires, having a seat designed to be straddled by the operator and handlebars for steering control. Their research reveals that TartanDrive data could be useful for training a self-driving vehicle to navigate off-road.


TartanDrive Data can Train All-Terrain Vehicles

Carnegie Mellon University researchers drove heavy equipment in all-terrain vehicles at 30 mph. They slid alternately, took it up and down hills, and even drove it in mud all while collecting data such as video, the speed of each wheel, and suspension from seven types of sensors. The resulting dataset is called TartanDrive.

TartanDrive is a massive data set with nearly 200,000 off-road interactions that may help future programmers understand physics so that vehicles can interpret terrain more intuitively. There is a genuine use for that in a rapidly changing world where infrastructural disasters can happen very suddenly.

Unlike autonomous street driving, off-road driving is more challenging because you have to understand the dynamics of the terrain to drive safely and to drive faster. Most of earlier work on rough terrain driving has been used for commenting on maps, which give marks like mud, grass, vegetation, or water to help the robot in figuring out its environmental elements. The team found that the multimodal sensor data they gathered from the TartanDrive empowered them to assemble expectation models predominantly.

The data that the research team assembled assisted them with building forecast models that worked better than models developed with simpler, non-dynamic data. By driving the all-terrain vehicle aggressively during tests, the team put the vehicle into a performance realm where an understanding of dynamics was essential. Robots that can comprehend elements are bound to have the option to reason about the actual world.