Thinklab-SJTU / Bench2Drive

[NeurIPS 2024 Datasets and Benchmarks Track] Closed-Loop E2E-AD Benchmark Enhanced by World Model RL Expert
Other
1.3k stars 85 forks source link

Question about traffic_light's 'affect_ego'. #28

Closed weimengchuan closed 4 months ago

weimengchuan commented 4 months ago

Hi, I have a question about traffic_light's 'affect_ego'.

I find that, in TJunction\TJunction_Town05_Route260_Weather0: frame 115 - 435, 'affect_ego' of all the traffic lights are set as False, but the red traffic light is just in front of ego car. Is that correct?

Thanks in advance!

Here is an example:

30 = {dict: 17} {'affects_ego': False, 'center': [46.06597137451172, -178.60494995117188, 3.0801031589508057], 'class': 'traffic_light', 'distance': 33.08744514532862, 'extent': [0.18995605409145355, 0.13933593034744263, 0.4345698356628418], 'id': '3286', 'lane_id': 1, 'location': [46.06597137451172, -178.60494995117188, 3.0801031589508057], 'road_id': 2043, 'rotation': [0.0, 0.0, 89.99988555908203], 'section_id': 0, 'semantic_tags': [6, 7], 'state': 2, 'trigger_volume_extent': [4.5, 0.5, 1.0], 'trigger_volume_location': [13.19943618774414, -190.031494140625, 1.0], 'trigger_volume_rotation': [0.0, 0.0, 89.99988555908203], 'type_id': 'traffic.traffic_light'}

31 = {dict: 17} {'affects_ego': False, 'center': [23.03403091430664, -215.04502868652344, 3.0801031589508057], 'class': 'traffic_light', 'distance': 39.796973984685536, 'extent': [0.18995849788188934, 0.13933593034744263, 0.4345698356628418], 'id': '3244', 'lane_id': -3, 'location': [23.03403091430664, -215.04502868652344, 3.0801031589508057], 'road_id': 2035, 'rotation': [0.0, 0.0, -89.9998779296875], 'section_id': 0, 'semantic_tags': [6, 7], 'state': 0, 'trigger_volume_extent': [4.5, 0.5, 1.0], 'trigger_volume_location': [55.00081253051758, -204.1710205078125, 1.0], 'trigger_volume_rotation': [0.0, 0.0, 270.0001220703125], 'type_id': 'traffic.traffic_light'}

32 = {dict: 17} {'affects_ego': False, 'center': [38.929386138916016, -157.58042907714844, 6.181065082550049], 'class': 'traffic_light', 'distance': 52.68453919433918, 'extent': [0.15461425483226776, 0.1771875023841858, 0.4355316162109375], 'id': '3239', 'lane_id': 1, 'location': [38.929386138916016, -157.58042907714844, 6.181065082550049], 'road_id': 2318, 'rotation': [0.0, 0.0, -4.999879837036133], 'section_id': 0, 'semantic_tags': [6, 7], 'state': 2, 'trigger_volume_extent': [3.0, 0.5, 1.0], 'trigger_volume_location': [36.155189514160156, -133.9046173095703, 1.2000017166137695], 'trigger_volume_rotation': [0.0, 0.0, 355.0001220703125], 'type_id': 'traffic.traffic_light'}

33 = {dict: 17} {'affects_ego': False, 'center': [23.94679069519043, -149.42938232421875, 6.081064701080322], 'class': 'traffic_light', 'distance': 65.79007838888423, 'extent': [0.1771887093782425, 0.15460936725139618, 0.4355316162109375], 'id': '3238', 'lane_id': -2, 'location': [23.94679069519043, -149.42938232421875, 6.081064701080322], 'road_id': 2307, 'rotation': [0.0, 0.0, -94.99996948242188], 'section_id': 0, 'semantic_tags': [6, 7], 'state': 0, 'trigger_volume_extent': [1.0, 0.5, 1.0], 'trigger_volume_location': [51.772605895996094, -149.50827026367188, 1.100000023841858], 'trigger_volume_rotation': [0.0, 0.0, 265.0000305175781], 'type_id': 'traffic.traffic_light'}

jiaxiaosong1002 commented 4 months ago

@weimengchuan It might be because some threshold issue. You may write your own rule to determine it if it is needed.