决策规划算法工程师（J93883）
Core
Design deep learning-based driving behavior decision models and trajectory planning algorithms to solve complex interaction problems like intersection navigation, lane-changing games, and yielding to pedestrians.
Role type
Senior IC autonomous driving decision planning algorithm engineer
Builds
End-to-end or hierarchical trajectory planning algorithms for autonomous vehicles
Domain
Autonomous driving, deep reinforcement learning, multi-agent systems
Deliverable
production ML models
Required skills
Deep reinforcement learning (DRL), Imitation Learning, Partially Observable Markov Decision Processes (POMDP), PyTorch, TensorFlow, Python, C++, Linux, Classical decision planning algorithms (A*, RRT*, MPC), Vehicle dynamics modeling
Preferred skills
Multi-agent systems, Game theory, Bayesian Networks, Top-tier conference publications (CoRL, ICRA, IV), Open-source project contributions (Apollo, Autoware), Robot path planning experience
Responsibilities
Design decision models for scene understanding and interaction intent; Research socially-compatible planning; Develop trajectory generation strategies; Explore real-time planning under uncertainty
Seniority
Senior, hands-on IC