DRL_based_SelfDrivingCarControl
DRL_based_SelfDrivingCarControl copied to clipboard
The simulator!
I am glad to work on your codes. While working, the simulator is not loading. Can you describe the simulator connections? Is it uses Socket? or it starts automatically with the python code that you've provided?
Thanks, Fayjie
Thanks for working on my code! It uses socketio to connect DQN code(Python) and Simulator(Unity). At first, execute python code (drive_Combined.py) Then download the simulator file via my link on the readme.md and execute simulator exe file. After that they connects each other and run!
I'm using your simulator for my project work, is there any way to cite it, also is there any name for this simulator??
@Kyushik
@matrixBT Sorry for late reply. I went to military training for a month, so I couldn't use any electronic devices. Actually, there is no cite or name for the simulator. I think adding github link is okay. Thanks for using my simulator for your project work!! Are you writing a paper with this simulator?
Excuse me , I can't find that 'drive_Combined.py'file in your git hub. Where shod I find that python file?
@champcui Hi! 'drive_combined.py' is a file that existed in old version. Now you should run the files in the 'RL_algorithms' folder
@champcui Hi! 'drive_combined.py' is a file that existed in old version. Now you should run the files in the 'RL_algorithms' folder
Thank u very much! I can use Rl_algorthms now, and I want to know how to get photo that in your paper just like "lanechange" "sensor.gif", and so on.
I recorded a video and converted it to gif :)
我录制了视频并将其转换为gif :)
okay, and how should I get that Input Configuration ?for example averge speed ,lanechange and number of overtaking. Because I can just get step and score in that algorthms.ipynb.
我录制了视频并将其转换为gif :)
It's much trouble for me if I record these every data in running video.
You can get speed and other data using vector observation in ipynb file :)
您可以使用ipynb文件中的矢量观测来获取速度和其他数据:)
Code for tensorboard tensorboard
def setup_summary(): episode_speed = tf.Variable(0.) episode_overtake = tf.Variable(0.) episode_lanechange = tf.Variable(0.)
tf.summary.scalar('Average_Speed/' + str(Num_plot_episode) + 'episodes', episode_speed)
tf.summary.scalar('Average_overtake/' + str(Num_plot_episode) + 'episodes', episode_overtake)
tf.summary.scalar('Average_lanechange/' + str(Num_plot_episode) + 'episodes', episode_lanechange)
summary_vars = [episode_speed, episode_overtake, episode_lanechange]
summary_placeholders = [tf.placeholder(tf.float32) for _ in range(len(summary_vars))]
update_ops = [summary_vars[i].assign(summary_placeholders[i]) for i in range(len(summary_vars))]
summary_op = tf.summary.merge_all()
return summary_placeholders, update_ops, summary_op
Sorry to bother you again! Actually, I find these code in your ipynb. But I can not find these calculate data at last. I realy want to use these data. Thanks!
I wrote information about vector observation on readme as follows. I think this info can help you :)
I know actions is related to max Qvalue. But I can't find actions and rewards in your ipynb. So how should I do to find it? Thanks for your paintience again!
Action is decided based on the neural network of the code and reward is returned as a result of the action as follows.
结果 Thanks! And I want to know how control your speed and lanechange in ipynb?
The speed and lanechange should be changed in the Unity project.