NGSIM_flexible_Testbed

A testbed for the autonomous driving controller/algorithms based on NGSIM datase

What can this testbed do

1. test your high-level decison-making algorithm (e.g. lane-change decision) with real traffic data
2. test your low-level dynamic controller (e.g. vehicle following controller) with real traffic data
3. visualize your results with the animations

What are already in the codes

1. NGSIM data processing part
2. Neighbor vehicles motion state tracking part (it can provide (x,y) informaton of the front car, left-front car, left rear car, right-front car, and right-rear car)
3. pre-defined vehicle dynamic model (bicycle model) Utils/vehicle_dynamics.py
4. pre-defined low-level controllers: vehicle speed controller, vehicle following controller, lane-keeping controller, and lane-changing controller Utils/low_level_controller.py
5. pre-defined psuedo-lane-change decision making part
6. animation part

How to play with this testbed

1. choose which car you are interested in (the testbed will duplicate a car operated under your algorithms) python data_processing.py
2. design/modify the high-level or low-level controller whichever you want
3. run the simulation python simulator.py

blue car is the original one (from NGSIM dataset) and red car is the duplicated car (with the controllers)