Skip to content
RobotWorld
← Back to Videos

SSIO: Self-Supervised Inertial Odometry Without Labels

Loading video
Loading video

SSIO removes the need for ground-truth labels by learning inertial odometry directly from raw IMU data, targeting robots, drones, and wearables without trajectory annotations.

SSIOInertial OdometrySelf-Supervised LearningIMURobot State EstimationLocalization
Category: perception
Author: @QiuYuhengQiu
Date: 2026-10-28T00:00:00
Duration: 49.233s
Source: https://x.com/QiuYuhengQiu/status/2115493554273595674
Reference: https://arxiv.org/abs/2610.25988

Related Videos

DeFM: A Depth Foundation Model for Robotics
2:17

DeFM: A Depth Foundation Model for Robotics

@robotsdigest · 24 views · 2026-10-28
Depth perceptionFoundation Models
Ego-OSCAR: open-source sub-$200 wearable for robot training data
0:26

Ego-OSCAR: open-source sub-$200 wearable for robot training data

@21nemus · 558 views · 2026-08-22
Ego-OSCARFPV Labs
Sony Unveils a Desktop Naked-Eye 3D Display
0:14

Sony Unveils a Desktop Naked-Eye 3D Display

@AUTOMATONJapan · 23 views · 2026-10-28
SonySpatial Display
XREAL AURA: Android XR Glasses with Gemini
0:16

XREAL AURA: Android XR Glasses with Gemini

@testingcatalog · 20 views · 2026-10-28
XREAL AURAAndroid XR
Hesai Kosmo: LiDAR + 8K Capture for Cloud 3D Reconstruction
0:36

Hesai Kosmo: LiDAR + 8K Capture for Cloud 3D Reconstruction

@tokufxug · 79 views · 2026-10-25
3D reconstruction3D Gaussian Splatting
Catching LiDAR Ghost Cars by Their Missing Shadow
0:33

Catching LiDAR Ghost Cars by Their Missing Shadow

@AlloyRobotics · 69 views · 2026-10-24
LiDARAdversarial Attack

As an Amazon Associate, we earn from qualifying purchases.