ICRA 2018 Workshop
May 25th
(Room M4 (Mezzanine level))
Embedded Files
http://longtermautonomy.eu
From driverless cars to autonomous harvesters to service robots, mobile robots are leaving the factories and entering less structured, more complex and dynamic environments. The key competence of these robots is the ability of reliable operation for long periods of time under changing and unpredictable environmental conditions. In other words, these robots need to be persistent and demonstrate a high level of robustness and fault tolerance and recovery, and above all of that, they have to be able to adapt over time to the changes in their operational environment. However, development and testing of these competences is difficult and tedious, because it goes far beyond running proof-of-concept experiments in controlled environments for a limited period of time. This makes any experience with long-term deployment of autonomous systems a valuable knowledge. In this workshop, we invite a number of renowned experts in the field who will highlight the main challenges these robots face and talk about their own experiences and the lessons they learnt during long-term deployments of their robots. We also call for papers that address the long-term autonomy problem and in particular the topics below:
From driverless cars to autonomous harvesters to service robots, mobile robots are leaving the factories and entering less structured, more complex and dynamic environments. The key competence of these robots is the ability of reliable operation for long periods of time under changing and unpredictable environmental conditions. In other words, these robots need to be persistent and demonstrate a high level of robustness and fault tolerance and recovery, and above all of that, they have to be able to adapt over time to the changes in their operational environment. However, development and testing of these competences is difficult and tedious, because it goes far beyond running proof-of-concept experiments in controlled environments for a limited period of time. This makes any experience with long-term deployment of autonomous systems a valuable knowledge. In this workshop, we invite a number of renowned experts in the field who will highlight the main challenges these robots face and talk about their own experiences and the lessons they learnt during long-term deployments of their robots. We also call for papers that address the long-term autonomy problem and in particular the topics below:
Schedule
Schedule
08:50 Organizers, Welcome and workshop overview
--
09:00 Cyrill Stachniss - Visual Localisation Under Strong Appearance Changes
09:30 Torsten Sattler - Visual Localisation Under Day-Night Changes
10:00 Carl Wellington - Long-term deployment of self-driving cars and trucks
--
10:30 Morning Tea
--
11:00 John Leonard - Long-term autonomy for Self-Driving Cars: Challenges and Opportunities
11:30 3 minutes lightning talks
- Amanda Mahony, Hadrien Bride, Jin Song Dong, Zhé Hóu, Brendan Mahony, Martin Oxenham: A Trusted Goal Reasoning and Planning Framework for Long Term Autonomy
- Hyungjin Kim and Hyun Myung: Efficient Map Management Scheme for LiDAR-based Vehicle Localization
- Titus Cieslewski and Davide Scaramuzza: Describe Structure, not Appearance: Place Recognition in Sparse or Semi-Dense Point Clouds that are Obtained from Vision Only
- Giseop Kim and Byungjae Park and Ayoung Kim: Learning Scan Context toward Long-term LiDAR Localization
- Younggun Cho, Jinyong Jeong, Youngsik Shin and Ayoung Kim: DejavuGAN: Multi-temporal Image Translation toward Long-term Robot Autonomy
- Marcin Dymczyk, Marius Fehr, Thomas Schneider and Roland Siegwart: Long-term Large-scale Mapping and Localization Using maplab
- Markus Bajones, Timothy Patten, Michael Zillich and Markus Vincze: Probabilistic Observation Maps for Use in Long-Term Human-Robot Interactions
--
12:00 Lunch and posters
--
14:00 Paul Newman - One Map to Rule Them All?
14:30 Joydeep Biswas - Deploying Mobile Robots and Keeping them Autonomous --
15:00 Coffee break and posters
--
16:00 Jonathan Kelly - Designing for Long-Term Autonomy: Experiences With Collaborative Robots
16:30 Tomas Krajnik - One Year of Autonomy in Everyday Environments: The STRANDS Project
--
17:00 Panel discussion
18:00 Concluding remarks
Workshop topics:
Workshop topics:
- Spatial and temporal representations for long-term mapping and localization,
- Reasoning about environmental appearance and structural change,
- Long-term mission planning and exploration,Lifelong learning and adaptation,
- Context-dependent decision making,
- Verification of long-term autonomous systems,
- Robust recovery behaviors.
Keynote speakers
Keynote speakers
- Carl Wellington, Uber Advanced Technologies Group
- Joydeep Biswas, UMass Amherst
- Paul Newman, University of Oxford
- Cyrill Stachniss, University of Bonn
- Jonathan Kelly, University of Toronto
- John Leonard, MIT
- Torsten Sattler, ETH
- Tom Krajnik, CTU
Organisers:
Organisers:
- Feras Dayoub, Australian Centre for Robotic Vision, Queensland University of Technology, AU
- Lars Kunze, Oxford Robotics Institute, Oxford, UK
- Tomas Krajnik, Czech Technical University in Prague
- Niko Suenderhauf, Australian Centre for Robotic Vision, Queensland University of Technology, AU
- Manuela Veloso, Carnegie Mellon University, USA
Programme committee:
Programme committee:
- Titus Cieslewski, University of Zurich
- Marcin Dymczyk, ETH Zurich
- Stephanie Lowry, Örebro University
- Peer Neubert, Technical University of Chemnitz
- Jaime Pulido Fentanes, University of Lincoln
- Joao Santos, FARO Labs
Important dates
Important dates
extended to April 8th Submission - follow the link
April 29th Notification of acceptance
May 6th Camera ready paper
May 25th Workshop (full-day)
Format:
Format:
Authors are required to submit a 2 pages extended abstract or 4 pages short paper or 6 pages full paper as PDF in the standard ICRA conference format.
Submissions:
Submissions:
https://easychair.org/conferences/?conf=icra18ltaw
Submissions will be judged based on relevance to the workshop topics, technical quality, and novelty. Authors of accepted papers are expected to give a lightning talk (2-3 minutes) and to present a poster at the workshop. A number of full papers will be selected for longer 10-15 oral presentation. The authors of the best full paper will be awarded 0.01 BTC.
Google Sites
Report abuse