Introduction

Key Responsibilities

Lead research and development of ​2D LiDAR-based SLAM algorithms for unknown environment mapping and self-localization in floor cleaning robots. Design and optimize ​motion distortion correction algorithms and ​multi-sensor fusion solutions (LiDAR/IMU/odometry). Develop and refine ​back-end optimization algorithms (e.g., Bayesian filtering, graph optimization, branch-and-bound methods) for robust SLAM performance. Innovate ​relocalization algorithms to recover from localization failures in dynamic environments. Research ​2D point cloud feature extraction methods, including clustering algorithms for structured/unstructured environments.

​Requirements

Technical Expertise:

Strong knowledge of ​LiDAR SLAM core components: scan matching, loop closure detection, Bayesian filtering, and map optimization.Hands-on experience with ​SLAM frameworks: Cartographer, KartoSLAM, or Gmapping.Proficiency in ​sensor integration: deep understanding of LiDAR/IMU/odometry principles and fusion techniques (Kalman filters, particle filters).

Development Skills:

Proficient in ​C++ and Python with clean coding practices.Experience with ​ROS/ROS2 for robotic system development.Demonstrated work on ​2D point cloud processing (feature extraction, descriptor matching).

Preferred Qualifications:

3+ years of experience in ​mobile robot SLAM development (household/service robots preferred).Publications or patents in SLAM-related fields.Fluency in spoken English for global team collaboration.

Education:

Bachelor’s/Master’s degree in ​Automation, Computer Science, Mathematics, or related fields.