Challenges in SLAM: What's ahead | Sebastian Scherer | Tartan SLAM Series скачать в хорошем качестве

Challenges in SLAM: What's ahead | Sebastian Scherer | Tartan SLAM Series

Presentation by Sebastian Scherer as part of the Tartan SLAM Series. Series overviews and links can be found on our webpage: https://theairlab.org/tartanslamseries/ This session gives an overview of the current SLAM systems, how they can be evaluated, and some of the current challenges. Outline: 0:00 - Welcome & Intro 5:36 - SLAM Resources 6:59 - Vernacular and Background 11:07 - Challenging Environments 12:03 - Current SLAM Systems 17:06 - Evaluating SLAM Systems 22:10 - State Estimation 28:30 - Localization 35:09 - Mapping 41:15 - Summary 42:41 - Open Discussion 44:46 - How are ground truths determined? 45:58 - Best SLAM for on edge device? 47:06 - Visual place recognition in SLAM? 48:56 - Internship opportunities? 49:59 - How measure SLAM performance? 51:00 - How tie SLAM closely to specific tasks? 54:12 - Visual SLAM for day and night? 55:24 - Sensors for snow and fog? 56:16 - How deal with brownout? 57:30 - When use deep learning or traditional methods? 1:00:24 - Are learning methods just used for feature extraction? 1:03:07 - Best way to fuse sensors? 1:04:48 - Active SLAM? 1:06:33 - How contribute to TartanAir dataset? AirLab Links: Website: https://theairlab.org Twitter:   / airlabcmu   LinkedIn:   / the-air-lab-at-carnegie-mellon-university   Facebook:   / airlabcmu   Medium:   / airlabcmu   RPL Links: Website: https://rpl.ri.cmu.edu/ Twitter:   / rpl_cmu