|
Marvin Chancán mchancanl at uni.pe I am completing a Ph.D. in Electrical Engineering and Robotics at Queensland University of Technology (QUT), Australia, advised by Prof. Michael Milford. Previously, I received an M.Sc. in Mechatronics Systems from PUC-Rio and a B.Sc. in Mechatronics Engineering (1st in my class) from UNI. I also worked in the automation and IT industries for over 6 years in several countries of the Americas and Europe. |
|
Google Scholar GitHub LinkedIn YouTube
Research
During my PhD, I focused on the intersection between deep learning, computer vision, neuroscience, and reinforcement learning with the aim of developing faster learning algorithms and techniques for localization and navigation tasks in mobile robots and autonomous vehicles. Specifically, I studied neural-level network architectures that underlie these complex tasks in insects (e.g. Drosophila melanogaster) and rat mammalian brains, and developed new neural network models that were found to be orders of magnitude faster and smaller than state-of-the-art deep learning systems.News
- Given the increasing interest on the CANN component of this paper, a quick demo has been released here.
- I have been accepted to the Janelia Research Campus's 2020 Junior Scientist MCN Workshop.
- 1 paper accepted to the NeurIPS 2020 Workshop on Machine Learning for Autonomous Driving.
- I am the recipient of the 2020 QUT HDR High Achiever Award.
- Joined the Neuro-Autonomy project. A US-Australia Multidisciplinary University Research Initiative (AUSMURI).
- 1 paper accepted at RSS 2020 Workshop on Self-Supervised Robot Learning.
- 2 papers accepted to ICRA 2020.
- First paper of my PhD accepted to IEEE Robotics and Automation Letters journal.
Publications
|
DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition Marvin Chancán, Michael Milford Machine Learning for Autonomous Driving Workshop at the 34th Conference on Neural Information Processing Systems (NeurIPS 2020) Project page | Video | Code |
|
Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning Marvin Chancán, Michael Milford Self-Supervised Robot Learning Workshop at the Robotics: Science and Systems XVI (RSS 2020) Project page | Video | ArXiv |
|
CityLearn: Diverse Real-World Environments for Sample-Efficient Navigation Policy Learning Marvin Chancán, Michael Milford IEEE International Conference on Robotics and Automation (ICRA 2020) Project page | Video | Code |
|
A Hybrid Compact Neural Architecture for Visual Place Recognition Marvin Chancán, Luis Hernandez-Nunez, A. Narendra, Andrew B. Barron, Michael Milford IEEE Robotics and Automation Letters (RA-L) Presented at ICRA 2020 Project page | Video | Code |
Preprints
|
MVP: Unified Motion and Visual Self-Supervised Learning for Large-Scale Robotic Navigation Marvin Chancán, Michael Milford ArXiv preprint 2020 Project page | ArXiv |
Research Experience
- PhD Student | AUSMURI, QUT Oct 2019 – Present
Project: Neuro-Autonomy: Neuroscience-Inspired Perception, Navigation, and Spatial Awareness for Autonomous Robots. - PhD Researcher | QUT Apr 2018 – Present
Topic: Brain-inspired localization and learning for robots and autonomous vehicles. - Research Assistant | PUC-Rio Aug 2012 – Oct 2012
Topic: Optimization of petroleum production process using genetic algorithms (GA). - Research Student | PUC-Rio Aug 2011 – Aug 2012
Topic (MSc dissertation): Adaptive control of intake air properties for engine testing using fuzzy logic. - Research Student | UNI Jun 2009 – Dec 2009
Topic (BSc thesis): Control of a three-phase induction motor via neural networks using a digital signal processor (DSP).
Service
- Journal Reviewer:
- IEEE Robot. Autom. Lett. 2020
- J. Braz. Soc. Mech. Sci. & Eng. 2019
- Int. J. Fuzzy Syst. 2018 - Conference Reviewer:
- IEEE/RSJ Int. Conf. Intell. Robot. & Syst. 2020
Honors, Awards and Scholarships
|