Skip to content
RobotWorld
← Back to Videos

FTP-1: A Generalist Foundation Tactile Policy

Loading video
Loading video

FTP-1 unifies image-, array-, and state-based tactile sensing through a Morphology-Aware Tactile Token Space, pretrained on about 3,000 hours of heterogeneous manipulation data.

Tactile SensingFoundation ModelsRobot ManipulationDexterous ManipulationVision-language-action (VLA)Cross-EmbodimentTactile PolicyMTTSFTP-1SharpaUniVTACContact-Rich ManipulationTactile Perception
Category: arm
Author: @LeoKharon
Date: 2026-09-14T00:00:00
Duration: 51.033s
Source: https://x.com/LeoKharon/status/2099468876618547388
Reference: https://ftp1-policy.github.io/

Related Videos

TacZero: Training-Free Peg Insertion with Tactile VLM
0:48

TacZero: Training-Free Peg Insertion with Tactile VLM

@sports_robots · 338 views · 2026-10-07
TacZeroVision-language models
DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation
2:25

DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation

@berkeley_ai · 650 views · 2026-09-29
Dexterous ManipulationTactile Sensing
Liber-0: Turning Human Experience into Robot Behavior
1:06

Liber-0: Turning Human Experience into Robot Behavior

@liberai_robot · 1.7k views · 2026-09-17
Dexterous ManipulationRobot Learning
DemoMimic: One Demo, Many Unseen Objects
0:40

DemoMimic: One Demo, Many Unseen Objects

@heetezition · 1.2k views · 2026-09-03
DemoMimicDexterous Manipulation
FACT: Learning Dexterous Manipulation from Video and Force
4:02

FACT: Learning Dexterous Manipulation from Video and Force

@RavenHuang4 · 73 views · 2026-10-28
FACTDexterous Manipulation
Wearable Data Collection Replaces Teleoperation
0:21

Wearable Data Collection Replaces Teleoperation

@NathanKau · 106 views · 2026-10-26
Embodied AIRobot Post-training

As an Amazon Associate, we earn from qualifying purchases.