← Back to Videos

ODE Policy: Continuous Action Fields for Robot Control

Loading video

NTU MARS open-sources ODE Policy, which models continuous action dynamics with an ODE vector field for smoother, low-latency robot control.

ODE Policy Robot policies Continuous Control Robot Manipulation NTU MARS

Category: arm

Author: @VedaAI00

Date: 2026-11-06T00:00:00

Duration: 90.09s

Source: https://x.com/VedaAI00/status/2118738212495798758

Two Models Attempt a Multi-Step Desk Organization Task

Two Models Attempt a Multi-Step Desk Organization Task

@embodianalysis · 45 views · 2026-11-05

RoboticsRobot Manipulation

I3L Turns Human Takeovers Into Better Robot Policies

I3L Turns Human Takeovers Into Better Robot Policies

@robotsdigest · 120 views · 2026-10-31

Imitation LearningInteractive Learning

Phone video to robot digital twin in under an hour

Phone video to robot digital twin in under an hour

@realfxw · 155 views · 2026-10-24

RoboSimStudioDigital twins

OneWM: World and Action Models That Strengthen Each Other

OneWM: World and Action Models That Strengthen Each Other

@Selen7005717917 · 164 views · 2026-10-24

World ModelsRobot Manipulation

3DPWM: Completed Point Clouds for Robot Planning

3DPWM: Completed Point Clouds for Robot Planning

@fuxinli2 · 184 views · 2026-10-20

3DPWMPoint Cloud Completion

ASPIRE: Model Frontier Meets Robot Cost

ASPIRE: Model Frontier Meets Robot Cost

@DrJimFan · 228 views · 2026-10-16

As an Amazon Associate, we earn from qualifying purchases.