Motional is a driverless technology company making autonomous vehicles a safe, reliable, and accessible reality. The Principal Engineer will work on the Spatial Intelligence Team to define and execute computer-vision based mapping projects that enhance self-driving vehicle capabilities and localize them using various data sources.
Responsibilities:
- Define, lead and execute computer-vision based mapping projects that improve our self-driving vehicles’ capability to efficiently map roadways and localize itself using vehicle and remote sensing data
- Develop methods to monitor changes to the environment and update maps accordingly
- Integrate visual maps with LIDAR-derived maps to create consistent fused world maps
- Productionize and deploy solutions onto autonomous vehicle fleets
- Collaborate with LIDAR-based localization and mapping, and perception teams to improve our vehicle’s on-road performance
Requirements:
- Experience developing software for geometric computer vision - geometric camera models, camera calibration, fundamental / essential matrices, the epipolar constraint, feature detection and matching, triangulation
- Experience with visual simultaneous localization and mapping (VSLAM), structure from motion, visual-inertial SLAM, visual-inertial odometry, and large-scale bundle adjustment/factor graph optimization at scale
- Proven track record of designing, developing and deploying 3D computer vision solutions for autonomous vehicles, augmented reality, robotics or related applications
- Masters or Ph.D. in Computer Science or a related technical field; or equivalent industry experience
- 7+ years of professional software engineering experience
- Advanced knowledge of software engineering principles including software design, source control management, build processes, code reviews, testing methods
- Extensive experience in metrics design and metrics driven technology development
- Excellent communication and interpersonal skills
- Strong programming skills in C++
- Proven track record of publications in relevant conferences (CVPR, ICML, NeurIPS, ICCV, WACV, AAAI, ICL, etc.)
- Experience with multi-modality, multi-sensor fusion-based mapping, e.g. visual + LIDAR-based mapping
- Remote sensing experience including geospatial coordinate frames, RPC camera models
- Strong python programming skills
- Experience training ML models using Pytorch or other libraries