Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach
Autonomous navigation of miniaturized robots (e.g., nano/pico aerial vehicles) is currently a grand challenge for robotics research, due to the need of processing a large amount of sensor data (e.g., camera frames) with limited on-board computational resources. In this paper we focus on the design o...
Main Authors: | , , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | en_US |
Published: |
2017
|
Online Access: | http://hdl.handle.net/1721.1/109522 https://orcid.org/0000-0002-0619-8199 https://orcid.org/0000-0002-0376-4220 https://orcid.org/0000-0003-1884-5397 https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-2225-7275 |
_version_ | 1826216240338698240 |
---|---|
author | Zhang, Zhengdong Suleiman, Amr AbdulZahir Carlone, Luca Sze, Vivienne Karaman, Sertac |
author2 | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
author_facet | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Zhang, Zhengdong Suleiman, Amr AbdulZahir Carlone, Luca Sze, Vivienne Karaman, Sertac |
author_sort | Zhang, Zhengdong |
collection | MIT |
description | Autonomous navigation of miniaturized robots (e.g., nano/pico aerial vehicles) is currently a grand challenge for robotics research, due to the need of processing a large amount of sensor data (e.g., camera frames) with limited on-board computational resources. In this paper we focus on the design of a visual-inertial odometry (VIO) system in which the robot estimates its ego-motion (and a landmark-based map) from on- board camera and IMU data. We argue that scaling down VIO to miniaturized platforms (without sacrificing performance) requires a paradigm shift in the design of perception algorithms, and we advocate a co-design approach in which algorithmic and hardware design choices are tightly coupled. Our contribution is four-fold. First, we discuss the VIO co-design problem, in which one tries to attain a desired resource-performance trade-off, by making suitable design choices (in terms of hardware, algorithms, implementation, and parameters). Second, we characterize the design space, by discussing how a relevant set of design choices affects the resource-performance trade-off in VIO. Third, we provide a systematic experiment-driven way to explore the design space, towards a design that meets the desired trade-off. Fourth, we demonstrate the result of the co-design process by providing a VIO implementation on specialized hardware and showing that such implementation has the same accuracy and speed of a desktop implementation, while requiring a fraction of the power. |
first_indexed | 2024-09-23T16:44:32Z |
format | Article |
id | mit-1721.1/109522 |
institution | Massachusetts Institute of Technology |
language | en_US |
last_indexed | 2024-09-23T16:44:32Z |
publishDate | 2017 |
record_format | dspace |
spelling | mit-1721.1/1095222022-09-29T21:12:31Z Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach Zhang, Zhengdong Suleiman, Amr AbdulZahir Carlone, Luca Sze, Vivienne Karaman, Sertac Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology. Microsystems Technology Laboratories Sze, Vivienne Zhang, Zhengdong Suleiman, Amr AbdulZahir Carlone, Luca Sze, Vivienne Karaman, Sertac Autonomous navigation of miniaturized robots (e.g., nano/pico aerial vehicles) is currently a grand challenge for robotics research, due to the need of processing a large amount of sensor data (e.g., camera frames) with limited on-board computational resources. In this paper we focus on the design of a visual-inertial odometry (VIO) system in which the robot estimates its ego-motion (and a landmark-based map) from on- board camera and IMU data. We argue that scaling down VIO to miniaturized platforms (without sacrificing performance) requires a paradigm shift in the design of perception algorithms, and we advocate a co-design approach in which algorithmic and hardware design choices are tightly coupled. Our contribution is four-fold. First, we discuss the VIO co-design problem, in which one tries to attain a desired resource-performance trade-off, by making suitable design choices (in terms of hardware, algorithms, implementation, and parameters). Second, we characterize the design space, by discussing how a relevant set of design choices affects the resource-performance trade-off in VIO. Third, we provide a systematic experiment-driven way to explore the design space, towards a design that meets the desired trade-off. Fourth, we demonstrate the result of the co-design process by providing a VIO implementation on specialized hardware and showing that such implementation has the same accuracy and speed of a desktop implementation, while requiring a fraction of the power. United States. Air Force Office of Scientific Research. Young Investigator Program (FA9550-16-1-0228) National Science Foundation (U.S.) (NSF CAREER 1350685) 2017-06-01T21:09:22Z 2017-06-01T21:09:22Z 2017-07 Article http://purl.org/eprint/type/ConferencePaper http://hdl.handle.net/1721.1/109522 Zhang, Zhengdong, Amr Suleiman, Luca Carlone, Vivienne Sze, Sertac Karaman. "Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach." Robotics: Science and System XIII, Cambridge, Massachusetts, 2017. https://orcid.org/0000-0002-0619-8199 https://orcid.org/0000-0002-0376-4220 https://orcid.org/0000-0003-1884-5397 https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-2225-7275 en_US http://rss2017.personalrobotics.ri.cmu.edu/program/papers/ Robotics: Science and Systems Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Sze |
spellingShingle | Zhang, Zhengdong Suleiman, Amr AbdulZahir Carlone, Luca Sze, Vivienne Karaman, Sertac Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title | Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title_full | Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title_fullStr | Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title_full_unstemmed | Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title_short | Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach |
title_sort | visual inertial odometry on chip an algorithm and hardware co design approach |
url | http://hdl.handle.net/1721.1/109522 https://orcid.org/0000-0002-0619-8199 https://orcid.org/0000-0002-0376-4220 https://orcid.org/0000-0003-1884-5397 https://orcid.org/0000-0003-4841-3990 https://orcid.org/0000-0002-2225-7275 |
work_keys_str_mv | AT zhangzhengdong visualinertialodometryonchipanalgorithmandhardwarecodesignapproach AT suleimanamrabdulzahir visualinertialodometryonchipanalgorithmandhardwarecodesignapproach AT carloneluca visualinertialodometryonchipanalgorithmandhardwarecodesignapproach AT szevivienne visualinertialodometryonchipanalgorithmandhardwarecodesignapproach AT karamansertac visualinertialodometryonchipanalgorithmandhardwarecodesignapproach |