New partnership with Rheinmetall to expand our autonomous navigation robustness

June 18th, 2020

Off-road terrains are hard to navigate due to the presence of a number of challenging and variable elements, such as mud, sand, rock, vegetation, and bodies of water on the ground. This variability in the terrain can hinder an autonomous vehicle’s ability to navigate with robustness.

Hence, off-road terrain navigation is a very challenging topic in robotics and requires further research. It has wide-reaching applications in a number of industries including agriculture, defence, mining, and construction to increase safety and productivity.

In partnership with Rheinmetall, an international group developing technologies in the mobility and security segments, we are working towards autonomously identifying which type of terrain a robotic vehicle is driving on, to increase its navigation capability.

For this task, our group is contributing our extensive experience in autonomous ground vehicle navigation, which will be combined with machine learning methods for terrain classification.

In this project we work in collaboration with the Imaging and Computer Vision Group, also part of the Cyber-Physical Research Program at CSIRO’s Data61, with their work focusing on the human activity recognition to assist autonomous vehicles in performing tasks.

Real-image (left), semantically segmented image (middle), overlay of segmentation and the real image, illustrating the results of terrain analysis technique.

We are open to partnerships and collaborations for research, innovation, development, and commercialisation.

For more information, contact us.


Subscribe to our News via Email

Enter your email address to subscribe and receive notifications of new posts by email.