Successfully reported this slideshow.

IMULet: A Cloudlet for Inertial Tracking

0

Share

Upcoming SlideShare
TARDEC Presentation 2
TARDEC Presentation 2
Loading in …3
×
1 of 45
1 of 45

More Related Content

Related Books

Free with a 14 day trial from Scribd

See all

IMULet: A Cloudlet for Inertial Tracking

  1. 1. IMULet: A Cloudlet for Inertial Tracking Lauri Tuominen and Mo Alloulah 25 Feb 2021 1 © Dania Alloulah
  2. 2. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet
  3. 3. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet
  4. 4. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet person chair chair robot drone bike person table Vision
  5. 5. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet Active RF 802.11mc person chair chair robot drone bike person table Vision
  6. 6. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet Active RF 802.11mc person chair chair robot drone bike person table Vision Passive RF-Pose3D
  7. 7. Navigation/Localisation is a fundamental technology Cornerstone of the physical internet Active RF 802.11mc person chair chair robot drone bike person table Vision Passive RF-Pose3D …
  8. 8. Multi-modal: SurroundSense Navigation/Localisation is a fundamental technology Cornerstone of the physical internet Active RF 802.11mc person chair chair robot drone bike person table Vision Passive RF-Pose3D …
  9. 9. Name of game: Robustification, robustification, robustification
  10. 10. The future of computing is spatial Name of game: Robustification, robustification, robustification
  11. 11. The future of computing is spatial multi-modal Name of game: Robustification, robustification, robustification
  12. 12. The future of computing is spatial multi-modal, multi-layered Name of game: Robustification, robustification, robustification
  13. 13. The future of computing is spatial multi-modal, multi-layered, and semantic Name of game: Robustification, robustification, robustification
  14. 14. The future of computing is spatial multi-modal, multi-layered, and semantic Lest … Name of game: Robustification, robustification, robustification
  15. 15. The future of computing is spatial multi-modal, multi-layered, and semantic Lest … Name of game: Robustification, robustification, robustification
  16. 16. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … …
  17. 17. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … …
  18. 18. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful
  19. 19. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful T r i c k s
  20. 20. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful T r i c k s DL
  21. 21. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful Comparison between DL (IONet), pedestrian dead reckoning (PDR), and traditional strapdown inertial navigation systems (SINS) Why Deep Learning? T r i c k s
  22. 22. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful T r i c k s Comparison between DL (IONet), pedestrian dead reckoning (PDR), and traditional strapdown inertial navigation systems (SINS) Chen et al. "Ionet: Learning to cure the curse of drift in inertial odometry." arXiv preprint arXiv:1802.02209 (2018). Why Deep Learning? T r i c k s
  23. 23. What do we need to do? Inertial Radar Lidar Vision Acoustic Infrared … … … … Ground 0 • independent of environmental dynamics • i.e. very useful T r i c k s Comparison between DL (IONet), pedestrian dead reckoning (PDR), and traditional strapdown inertial navigation systems (SINS) Chen et al. "Ionet: Learning to cure the curse of drift in inertial odometry." arXiv preprint arXiv:1802.02209 (2018). Why Deep Learning? T r i c k s Major improvements over classical methods
  24. 24. DL, is that it?
  25. 25. DL, is that it? System-level innovations needed for the real-world
  26. 26. DL, is that it? System-level innovations needed for the real-world • ultra-low power tags • compute-, energy-, and form factor-constrained
  27. 27. DL, is that it? System-level innovations needed for the real-world • ultra-low power tags • compute-, energy-, and form factor-constrained • deal with DL model fragility • signal characteristic variabilities in the field • i.e. hard to generalise
  28. 28. DL, is that it? System-level innovations needed for the real-world • ultra-low power tags • compute-, energy-, and form factor-constrained • deal with DL model fragility • signal characteristic variabilities in the field • i.e. hard to generalise • configurability and scalability • use cases/market segments
  29. 29. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s
  30. 30. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s
  31. 31. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s
  32. 32. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s
  33. 33. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (1) ultra-low power
  34. 34. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (1) ultra-low power
  35. 35. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (2) configurable
  36. 36. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (2) configurable
  37. 37. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (3) customisable D1 D2 D3 D4 D5 D6 D1 -91.3 -38.3 -37.3 -33 -34.6 -31 D2 -38.3 -91.6 -38.3 -32.6 -36.3 -30.8 D3 -37.3 -38.3 -91.2 -32.4 -35.8 -30.8 D4 -33 -32.6 -32.4 -91.8 -29.8 -28.9 D5 -34.6 -36.3 -35.8 -29.8 -91.8 -30.6 D6 -31 -30.8 -30.8 -28.9 -30.6 -91.2
  38. 38. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s (3) customisable D1 D2 D3 D4 D5 D6 D1 -91.3 -38.3 -37.3 -33 -34.6 -31 D2 -38.3 -91.6 -38.3 -32.6 -36.3 -30.8 D3 -37.3 -38.3 -91.2 -32.4 -35.8 -30.8 D4 -33 -32.6 -32.4 -91.8 -29.8 -28.9 D5 -34.6 -36.3 -35.8 -29.8 -91.8 -30.6 D6 -31 -30.8 -30.8 -28.9 -30.6 -91.2
  39. 39. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s Promising early performance
  40. 40. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s Promising early performance
  41. 41. IMULet LSTM stack states estimates delayline Edge Cloud raw 9-DOF IMU signals acceI gyro magneto Displacement Estimate On-device compressed representation FC FC Adaptation z K layer 1 CNN layer 0 Latent space tracking s Promising early performance Check out paper for details
  42. 42. In Summary IMULet is a cloudlet for inertial tracking • DL + edge + hooks for efficiency and generalisability
  43. 43. In Summary IMULet is a cloudlet for inertial tracking • DL + edge + hooks for efficiency and generalisability Key component towards infrastructure-less localisation • as mandated by physical internet/spatial computing
  44. 44. In Summary IMULet is a cloudlet for inertial tracking • DL + edge + hooks for efficiency and generalisability Key component towards infrastructure-less localisation • as mandated by physical internet/spatial computing Scaling evaluation in-progress
  45. 45. Cheers 45 © Dania Alloulah

×