Research on autonomous vehicle has mainly focused on 3 main layers: perception, planning, and control. Each one of these layers presents very different specificities, and are often addressed separately, with in general strong assumptions on the behavior of the other components of the vehicle.

1. Simultaneous Localization and Mapping

2. GLobal Planning (Model Predictive Control)

3. Localization (SynPF)

4. Local Cost Map

5. Controller (Model Path Integral Control)

kinematics에 pacejka magic fomula를 추가한 dynamics를 풀면 더 성능이 좋아지지 않을까?

차량의 더 복잡한 dynamics에는 어떤 게 있을까?

6. System

Reference

• The Kinematic Bicycle Model: a Consistent Model for Planning Feasible Trajectories for Autonomous Vehicles?