辅助驾驶感知融合工程师

该职位来源于猎聘 Key Tasks

• Design and iterate multi-modal fusion algorithms for camera, millimeter-wave radar and etc. Implement core functions including 3D object detection, multi-object tracking, classification, velocity & pose estimation, and environmental semantic modeling.
• Responsible for the overall system architecture design, interface definition and data flow optimization of the perception fusion module. Establish standard pipelines for sensor time synchronization, coordinate unification, cross-sensor object association and fusion confidence evaluation.
• Analyze massive real road test data, identify and resolve core perception issues including false detection, missing detection, object jitter and trajectory drift under challenging scenarios. Complete full-loop algorithm tuning and strategy iteration to improve the stability, accuracy and consistency of 360° full-scene perception for mass-production vehicles. Profile:
• Master’s degree or above in electrical engineering, robotics engineering, computer science, automotive, mechanical engineering
• Proficient in core fusion theories including Kalman Filter, data association, Hungarian algorithm, Bayesian estimation and multi-object tracking. Solid knowledge of 3D geometry, coordinate transformation.
• Proficient in C++ and Python programming. Familiar with Linux development environment, CMake compilation and debugging tools.
• Practical experience with Sparse4D, BEV Fusion, Memory Bank and other state-of-the-art temporal perception fusion frameworks
• Strong analytical and problem-solving skills,
• Strong sense of ownership and accountability, teamwork
• Fluent English oral and written skills to communicate with global algorithm team