{
  "id": 177289,
  "title": "About autonomous vehicles",
  "url": "/competitions/lyft-motion-prediction-autonomous-vehicles/discussion/177289",
  "author_name": "",
  "post_date": "2020-08-25T11:40:07.060750900Z",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi guys, Autonomous Vehicles is a very interesting problem so I guess many people including me would like to take this as an opportunity to learn and practice on the subject.</p>\n<p>But I have a question for people with more experience on this problem, is this challenge related to just a small subset of ML applied to autonomous driving, or does it gives a good broad overview? also, would it be only really useful to Lyft data/problems or to the field as a whole?</p>",
  "messages": [
    {
      "id": "984949",
      "postDate": "08/25/2020 11:40:07",
      "content": "<p>Hi guys, Autonomous Vehicles is a very interesting problem so I guess many people including me would like to take this as an opportunity to learn and practice on the subject.</p>\n<p>But I have a question for people with more experience on this problem, is this challenge related to just a small subset of ML applied to autonomous driving, or does it gives a good broad overview? also, would it be only really useful to Lyft data/problems or to the field as a whole?</p>",
      "rawMarkdown": "Hi guys, Autonomous Vehicles is a very interesting problem so I guess many people including me would like to take this as an opportunity to learn and practice on the subject.\n\nBut I have a question for people with more experience on this problem, is this challenge related to just a small subset of ML applied to autonomous driving, or does it gives a good broad overview? also, would it be only really useful to Lyft data/problems or to the field as a whole?",
      "votes": null
    },
    {
      "id": "985255",
      "postDate": "08/25/2020 15:28:16",
      "content": "<p>Motion prediction is only a part of an AV system.</p>\n<p>Roughly it looks like this:</p>\n<ul>\n<li><strong>Perception</strong> (works with raw data; radar, lidar, camera; lots of classification, detection, sensor fusion, etc)</li>\n<li><strong>Prediction</strong> For this competition, we are using the output of the perception</li>\n<li><strong>Planning</strong> After we predicted the future trajectories for the surrounding agents, the next step is motion planning</li>\n<li><strong>Control</strong> The last step is controlling (accelerate/brake, steering)</li>\n</ul>\n<p>Take a look at <a href=\"https://eng.uber.com/machine-learning-model-life-cycle-version-control/\" target=\"_blank\">this post</a> from Uber</p>",
      "rawMarkdown": "Motion prediction is only a part of an AV system.\n\nRoughly it looks like this:\n- **Perception** (works with raw data; radar, lidar, camera; lots of classification, detection, sensor fusion, etc)\n- **Prediction** For this competition, we are using the output of the perception\n- **Planning** After we predicted the future trajectories for the surrounding agents, the next step is motion planning\n- **Control** The last step is controlling (accelerate/brake, steering)\n\nTake a look at [this post](https://eng.uber.com/machine-learning-model-life-cycle-version-control/) from Uber",
      "votes": null
    },
    {
      "id": "985423",
      "postDate": "08/25/2020 18:02:47",
      "content": "<p>Thanks <a href=\"https://www.kaggle.com/pestipeti\" target=\"_blank\">@pestipeti</a> , the post seems a great introduction to the field, I will take a look for sure!</p>",
      "rawMarkdown": "Thanks @pestipeti , the post seems a great introduction to the field, I will take a look for sure!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 985255,
      "author_name": "pestipeti",
      "author_url": "",
      "post_date": "08/25/2020 15:28:16",
      "content": "<p>Motion prediction is only a part of an AV system.</p>\n<p>Roughly it looks like this:</p>\n<ul>\n<li><strong>Perception</strong> (works with raw data; radar, lidar, camera; lots of classification, detection, sensor fusion, etc)</li>\n<li><strong>Prediction</strong> For this competition, we are using the output of the perception</li>\n<li><strong>Planning</strong> After we predicted the future trajectories for the surrounding agents, the next step is motion planning</li>\n<li><strong>Control</strong> The last step is controlling (accelerate/brake, steering)</li>\n</ul>\n<p>Take a look at <a href=\"https://eng.uber.com/machine-learning-model-life-cycle-version-control/\" target=\"_blank\">this post</a> from Uber</p>",
      "votes": null,
      "replies": [
        {
          "id": 985423,
          "author_name": "dimitreoliveira",
          "author_url": "",
          "post_date": "08/25/2020 18:02:47",
          "content": "<p>Thanks <a href=\"https://www.kaggle.com/pestipeti\" target=\"_blank\">@pestipeti</a> , the post seems a great introduction to the field, I will take a look for sure!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "984949": "Hi guys, Autonomous Vehicles is a very interesting problem so I guess many people including me would like to take this as an opportunity to learn and practice on the subject.\n\nBut I have a question for people with more experience on this problem, is this challenge related to just a small subset of ML applied to autonomous driving, or does it gives a good broad overview? also, would it be only really useful to Lyft data/problems or to the field as a whole?",
    "985255": "Motion prediction is only a part of an AV system.\n\nRoughly it looks like this:\n- **Perception** (works with raw data; radar, lidar, camera; lots of classification, detection, sensor fusion, etc)\n- **Prediction** For this competition, we are using the output of the perception\n- **Planning** After we predicted the future trajectories for the surrounding agents, the next step is motion planning\n- **Control** The last step is controlling (accelerate/brake, steering)\n\nTake a look at [this post](https://eng.uber.com/machine-learning-model-life-cycle-version-control/) from Uber",
    "985423": "Thanks @pestipeti , the post seems a great introduction to the field, I will take a look for sure!"
  },
  "source": "meta"
}