shuijing725 / crowdnav_prediction_attngraph Goto Github PK

Hi! @Shuijing725 , i read your reference.
But there is no script about how to make train dataset.
Could you tell me how to create dataset?
(In my situation, actually i use other platform about DRL environment. )

How can I get more information about humans, such as vx, vy, course angle?

Dear researcher,

Thank you for your fantastic work. I am working on running your project. However, the messages received by /map seems not meaningful. In a map that is almost empty, the map.data I received is almost full of -1. If I understand it correctly, value 0 stands for open space without obstacles. So what is the meaning of -1 and 100:

Hi researchers,

Thank you for your elegant work. I am working on a project that requires a deeper dive into prediction models, and I am having some difficulty finding the prediction horizons, intervals between prediction steps, and intervals between simulation steps. Would you mind telling me the values or indicating which files contain these parameter settings? (I suppose if I use pretrained GST models, the prediction horizons and prediction frequency are set and cannot be changed by config files, correct?)

Thank you so much!

How to add static obstacles to the scene?

Hello, I really like your work！I have encountered some questions and would like to ask you.
Recently, I hope to transform one intelligent agent into multiple intelligent agents.

I hope these multiple agents adopt the same reinforcement learning strategy and set sim. predict_ method='Inferred'.

I would like to know which code needs to be modified?
Thank you.

Hi, please could you provide information about what compute resources you used to train the model and how long it took.

Hello,

I encounter a question for ros_turtlebot2i_env.
In the code

humanStatesSub = Subscriber('/dr_spaam_detections', PoseArray) # human px, py, vi

how is this topic published in the ROS work space?

Thanks.

"I have a question. How can I implement the images of Social Force and ORCA from the URL https://sites.google.com/view/intention-aware-crowdnav/home? I also want to see the effects of these two policies. Since I'm a beginner in this field, I don't know how to implement it."

Excuse me，may I ask what part of the crowd preprocessing module is that?

When I finished executing test.py, I failed to find the saved test results in the trained_models/your_output_dir/test/ folder
Hope you can give me some advice thanks

Hi @Shuijing725 !.
I have a issue about your code in training with srnn.
I changed robot.policy in config.py from "selfAttn_merge_srnn" to "srnn".
Then when i run the train.py, the error occurred like under line.

`<Monitor<CrowdSimPredRealGST>>
No ghost version.
new gst
new st model
LOADED MODEL
device: cuda:0

Traceback (most recent call last):
File "/root/dsrnn_ws/CrowdNav_Prediction_AttnGraph/train.py", line 247, in
main()
File "/root/dsrnn_ws/CrowdNav_Prediction_AttnGraph/train.py", line 90, in main
actor_critic = Policy(
File "/root/dsrnn_ws/CrowdNav_Prediction_AttnGraph/rl/networks/model.py", line 28, in init
self.base = base(obs_shape, base_kwargs)
File "/root/dsrnn_ws/CrowdNav_Prediction_AttnGraph/rl/networks/srnn_model.py", line 378, in init
robot_size = 7 if args.env_type == 'crowd_sim' else 2
AttributeError: 'Namespace' object has no attribute 'env_type'. Did you mean: 'env_name'`?

Could you solve this issue? Thank you for reading!

As the number of tests increases, the program runs longer and longer. Is there a memory leak?

Hi @Shuijing725 !
I read your paper and code. it was awesome!!.

Anyway, i have a single question.
I want to use your old version =DS-RNN-CrowdNav in this recently version.
In your config , there is exist about robot.policy.
So i modified this part to "srnn", and i tested by a example weight about recently version.
Did i run original version or recently version?

Hello, do you still remember me? I still have some issues with the test.py, when I set save_slides=True, the folder trained_models/GST_predictor_rand/social_eval/41665 where the results are saved is empty. There is also a problem with the --test_case parameter setting. When I set any other positive number, the test will still run for 500 episodes. I'm confused and made some modifications to evaluation.py but it doesn't work. Do you have any suggestions? Maybe the uploaded file is incomplete