The source code for our paper: Graph Reinforcement Learning Application to Co-operative Decision-Making in Mixed Autonomy Traffic: Framework, Survey, and Challenges. This code is developed based on our previous repository TorchGRL.
Hello, when I reproduced your code, I found that the vehicle lane change process is very fast in sumo, and it feels like the lane change is performed instantaneously. I think this is not in line with common sense in the real scene, so I would like to ask you if Is there something wrong with my related settings, or do you have some other meanings here? Looking forward to your reply, thank you~
The questions have been solved. The new folder named 'GRL_Trained_Models' should be pre-created and located above the folder containing 'main_HighwayRamps.py'. Thus, 'GRL_Trained_Models' would get the results.
Could you please share how to generate training plot such as reward curve? Need a Tutorial. Thanks.
I have run the highway simulation, and got GRL_TrainedModels/PPO/DQN5/Rewards.npy. But I don't know how to use your Data_Plot_Train.py to plot? Thank you for your guidance!
I have google the paper, " Graph Reinforcement Learning Application to Co-operative Decision-Making in Mixed Autonomy Traffic: Framework, Survey, and Challenges." . But I didn't find this paper. Where can I get this paper, or can you send it to me? My Email is : [email protected] . Thanks!
Thanks very much for your sharing. When I run your codes, the terminal shows the below warning:
Warning: Vehicle 'flow_2.2'; junction collision with vehicle 'flow_1.1', lane=':highway_1_1_0', gap=-1.00, time=12.00 stage=laneChange.
Warning: Vehicle 'flow_2.2' performs emergency braking with decel=-9.00 wished=4.50 severity=1.00, time=12.10.
May I ask if you also encounter these warnings when running the code and whether the warnings will affect the Graph_CAV's performance?
Thank you very much, I installed the process from the source code at the end of the first process during the trial installation:
python setup.py develop prompts an error error: docutils 0.16 is installed but docutils<0.20,>=0.18 is required by {'sphinx'}
Should this be resolved?