Autonomous navigation in the (animal and the) machine


Information about my PhD project and the [neoRL] framework for autonomous navigation.

View My GitHub Profile

[ neoRL ] [ demonstration ] [ about the author ]

Per R. L.

Per R. Leikanger

Autonomous researcher on autonomous navigation. My primary aim is to understand learning and autonomy to a level by which an autonomous agent can be implemented. My approach is to combine modern RL from AI with modern results from the neuroscience of navigation. See my publications for how autonomous navigation is possible in technology.

Please do not hesitate to reach out: per email or via LinkedIn.

2011 — Kavli Institute for Systems Neuroscience/ Biology of Memory, NTNU

In 2011 I had the pleasure of studying under May-Britt and Edvard Moser, years before the Mosers shared the 2014 Nobel Prize in Physiology and Medicine with John O’Keefe.
This was done on the side of completing my MSc degree in Engineering Cybernetics.

2012 — Engineering Cybernetics: Biomedical Robotics, NTNU

Master’s degree at Norwegian University of Technology and Science (NTNU).
“Development and assessment of a novel model for artificial neural networks”: MSc thesis in engineering cybernetics development and assessment of a novel model for the signalling properties of the neuron.

2012 - 2016 — R&D Engineer in the industry
2017 - 2022 — PhD: Autonomous Navigation by (the animal and) the machine.

A deep investigation in the discrepancies between digital “autonomy” and biological agents for autonomous navigation uncovered how genuine autonomy is plausible via distributed approaches to RL. Purposive networks, where desires formed based on experience can be used as setpoints for other neoRL modules, allows for genuine category II autonomous navigation. The AI equivalent to neobehaviorism and Tolman’s purposive behaviorism.

First half of 2020: Academic visitor with Richard Sutton at RLAI lab @ UoA.