Optimization Methods Developer for the Lidar Map Team

Autonomous vehicles must always know their precise location in space. One of the most reliable methods to solve this task is localization using a lidar map: a three-dimensional reconstruction of the world as a point cloud.

The lidar map development team works on mechanisms for the automated creation and updating of the lidar map, as well as global optimization algorithms for point clouds that form the basis of these mechanisms.

Tasks that await you

Development and support of map creation technologies We want to cover tens of thousands of kilometers with a lidar map, which will require a qualitative leap in the current creation technology. Come improve our mathematics and optimize the map creation code.

Detection of map obsolescence and its automated rebuilding We want to be able to quickly and timely detect when a part of the map has become obsolete and rebuild it.

Development of optimization algorithms We are solving an interesting mathematical problem with very large data. We have many complex and interesting tasks for programmers and computational mathematicians.

We expect you to

• Be proficient in C++ and Python
• Have used cloud data orchestration tools (Yandex Nirvana, Airflow, Dagster)
• Have worked with big data
• Have a good mathematical background and be ready to dive into the world of state-of-the-art optimization algorithms

Will be a plus

• Have solved SLAM/optimization tasks on factor graphs
• Know what ICP and NDT are