Local Navigation and Docking of an Autonomous Robot Mower using Reinforcement Learning and Computer Vision
Authors:
Ali Taghibakhshi,
Nathan Ogden,
Matthew West
Abstract:
We demonstrate a successful navigation and docking control system for the John Deere Tango autonomous mower, using only a single camera as the input. This vision-only system is of interest because it is inexpensive, simple for production, and requires no external sensing. This is in contrast to existing systems that rely on integrated position sensors and global positioning system (GPS) technologi…
▽ More
We demonstrate a successful navigation and docking control system for the John Deere Tango autonomous mower, using only a single camera as the input. This vision-only system is of interest because it is inexpensive, simple for production, and requires no external sensing. This is in contrast to existing systems that rely on integrated position sensors and global positioning system (GPS) technologies. To produce our system we combined a state-of-the-art object detection architecture, You Only Look Once (YOLO), with a reinforcement learning (RL) architecture, Double Deep QNetworks (Double DQN). The object detection network identifies features on the mower and passes its output to the RL network, providing it with a low-dimensional representation that enables rapid and robust training. Finally, the RL network learns how to navigate the machine to the desired spot in a custom simulation environment. When tested on mower hardware, the system is able to dock with centimeter-level accuracy from arbitrary initial locations and orientations.
△ Less
Submitted 2 March, 2021; v1 submitted 15 January, 2021;
originally announced January 2021.
Precise Near-Infrared Radial Velocities
Authors:
Peter Plavchan,
Peter Gao,
Jonathan Gagne,
Elise Furlan,
Carolyn Brinkworth,
Michael Bottom,
Angelle Tanner,
Guillem Anglada-Escude,
Russel White,
Cassy Davison,
Sean Mills,
Chas Beichman,
John Asher Johnson,
David Ciardi,
Kent Wallace,
Bertrand Mennesson,
Gautam Vasisht,
Lisa Prato,
Stephen Kane,
Sam Crawford,
Tim Crawford,
Keeyoon Sung,
Brian Drouin,
Sean Lin,
Stephanie Leifer
, et al. (9 additional authors not shown)
Abstract:
We present the results of two 2.3 micron near-infrared radial velocity surveys to detect exoplanets around 36 nearby and young M dwarfs. We use the CSHELL spectrograph (R ~46,000) at the NASA InfraRed Telescope Facility, combined with an isotopic methane absorption gas cell for common optical path relative wavelength calibration. We have developed a sophisticated RV forward modeling code that acco…
▽ More
We present the results of two 2.3 micron near-infrared radial velocity surveys to detect exoplanets around 36 nearby and young M dwarfs. We use the CSHELL spectrograph (R ~46,000) at the NASA InfraRed Telescope Facility, combined with an isotopic methane absorption gas cell for common optical path relative wavelength calibration. We have developed a sophisticated RV forward modeling code that accounts for fringing and other instrumental artifacts present in the spectra. With a spectral grasp of only 5 nm, we are able to reach long-term radial velocity dispersions of ~20-30 m/s on our survey targets.
△ Less
Submitted 18 March, 2016;
originally announced March 2016.
