fangjianli Goto Github PK

deep-learning-project-generate-pics-with-gans

deep-learning-project1-image-classification-with-dnn-cnn

deep-learning-project2-video-captioning-with-seq2seq

driving_simulator_for_rl_il

A driving simulator built specifically for reinforcement learning/imitation learning. The scenario generator is based on highway-env, and 3-D animation is based on Carla

fangjianli.github.io

A beautiful, simple, clean, and responsive Jekyll theme for academics

human_centered_cacc

The codes for the paper ``Cooperative adaptive cruise control for string stable mixed traffic: Benchmark and human-centered design" (2017)

ngsim_flexible_testbed

A testbed for the digital twin with autonomous driving controller/algorithms based on NGSIM dataset

ppo-and-airl-with-parallel-sampling

RL algorithm PPO and IRL algorithm AIRL written with Tensorflow. They have a parallel sampling feature in order to increase computation speed (especially in high-performance computing (HPC)).

safe_imitation_learning

Safe adversarial inverse reinforcement learning/imitation learning in the paper "Safety-Aware Adversarial Inverse Reinforcement Learning (S-AIRL) for Highway Autonomous Driving"

tzmg_mutiple_cars

Two play zero sum game framework is used to train the safe driving policy considering the maximum uncertainties of the neighbor cars