Hello, Thank you for your great work if I generate the dataset correctly, what will be its folder placement structure? Could you give me an example?Just a screenshot is ok.

Hi, Thanks for the amazing work! I followed the Data Generation section in README.md(I did not modify any configurations and just activated the export of SAVE_PATH in the script). The following video shows the generated front images:

| As shown in the above video, the weather is changing, and this example(as one RouteScenario) consists of 94 frames (Different scenarios have different frames).

For some scenarios, they met the failures like the followings:

Does it work? Is there anything wrong with this?

Hello, It seems that You sampled 64 points for 5 different categories during data set preprocessing, but in fact, the semantic segmentation in Bev is a dense prediction task. For example, in a range of 200 * 200 pixels, 40000 points need to be sampled, so how to process and visualize the results in the inference stage?

Hi, Thanks for the interesting work and well documentation!

I follow the instructions provided in the README file to run the Autopilot but I face the following errors:

1. I got this error (pic below) when I tried to run "run_evaluation.sh", I solved this problem by creating a directory named "carla_results". Is this the right solution?

2. Now, I'm getting another error (pic below), what can be the problem?

Any tips would be helpful.

Thanks in advance!

Thank you for providing a good code. Where can I download dataset for training? I referred to the config.py, but it seems to be different from the transfuser dataset. Transfuser provides Town0x_long, Town0x_short, and Town0x_tiny but it requires Town0x, Town0x_long, and Town0x_small.

Hi @kashyap7x , thanks for your wonderful work on NEAT. I would like to ask regarding the blobby nature of BEV output as compared to other BEV methods like LSS or Fiery, where BEV of vehicles is quite sharp. Can you tell some pointers on how we can improve this irregular vehicle bev coming in NEAT?

Thanks

Hello, Thanks for your wonderful work! I have doubt with the data generation, Is the dataset in this work the same as in transfuser? Shall I download the data in https://s3.eu-central-1.amazonaws.com/avg-projects/transfuser/data/14_weathers_minimal_data.zip and directly use it? And what's the data format for query_points? just simple (x,y,t)->(2,5,1) as in the paper described? If the semantic map range is huge, did you normalize the input data?

Thanks for the work and the open source code. In the controller, I know that the use of red light for the longitudinal controller. But what if just use raw data input of RGB to identify instead of semantic? Could you please tell me the affect of the semantic? That is I think the semantic is just use to identify the traffic light.

Hi, I am taking a look at your code and I noticed you predict 5 waypoints in the decoder network, but you only use 4 waypoints for actually controlling the car. You do not use the first predicted waypoint and I was wondering why you would do that.

Thanks in advance!

I am not able to replicate the gif shown in README wherein you are able to visualize waypoint offsets too along with dashcam and semantic segmentation from BEV.

Could you please share the scripts for the same?

Hi, Thanks for the interesting work !

I follow the step of Data Generation provides in the README file, "./leaderboard/scripts/run_evaluation.sh" to run the Autopilot. But I havesome questions,as the following : (1) this process is only to evaluate the performance of different agents on the defined routes. No data was generated. (2) different models (agent) need different types of data. I want to know how to generate training data for different models.

Any tips would be helpful.

Thanks in advance!

Dear All,

I am a student working on a project involving testing ADAS/AV and scenario synthesis using BeamNG.tech and would love to run (test) your driving agent in that simulator. I know CARLA is kind of a de facto standard, but IMHO BeamNG.tech is superior when it comes to physic simulation, content, and flexibility. Further, BeamNG.tech is free for research, offers a python API, just like CARLA, and implements a wide range of sensors.

So I wonder how technically difficult it would be to port NEAT to BeamNG.tech and whether anyone of you could support me (and my colleagues) in doing so. Hope to hear from you soon,

Thanks!

-- Benedikt Steininger

when i Running the Autopilot

./leaderboard/scripts/run_evaluation.sh

i accept this error :

Traceback (most recent call last): File "leaderboard/leaderboard/leaderboard_evaluator.py", line 477, in main leaderboard_evaluator.run(arguments) File "leaderboard/leaderboard/leaderboard_evaluator.py", line 406, in run self.statistics_manager.clear_record(args.checkpoint) File "/home/ev/Net/carla_prj/neat/leaderboard/leaderboard/utils/statistics_manager.py", line 340, in clear_record with open(endpoint, 'w') as fd: FileNotFoundError: [Errno 2] No such file or directory: '../carla_results/auto_pilot.json' Exception ignored in: <function LeaderboardEvaluator.del at 0x7fe8e9a8a9e0>

can you help? thank you very much