All Episodes
Embodied AI 101 — 387 episodes · page 2 of 4
Seeing the Nearly Invisible: Stereo Vision for Cable-Safe Agricultural UAVs
Taming the Flying Tip: Constrained MPC for Dynamic Deformable Linear Objects
Replay Is Not Reality: Embodied RL When Both Policy and Physics Shift
PPO Beyond the Simulator: What a TurtleBot3 Mapless Navigation Study Really Demonstrates
VLAs in the Wild: What 142 Papers Really Say About Generalist Robot Intelligence
VEDA and the Moment Vision Stops Being Enough
Three Ways to Buy Back Robot Data: Augment the Demonstration, Import the Semantics, Benchmark the Physics
UI-Mate: Show the Desktop Agent Once, Then Let It Replan
The Frozen VLM That Learns Anyway: Inside Spatial Memory Agent
The LLM Proposes, the Planner Disposes: VLA-SP for Verifiable Robot Manipulation
When Vision Cannot Know: MulPlanLM Lets Robot Planners Lift, Weigh, and Decide
Reading Trust in Motion: A Multimodal Perception Stack for Industrial HRC
Two MPCs, One Robot: Bézier Curves for Fast, Compliant Mobile Manipulation
The Reward Behind the Gait: Learning Transferable Locomotion from a Few Stick-Insect Steps
Beyond Looking Human: Bio-Functional Robot Design for Cross-Scale Dexterity and Tool Use
LDA-1B and the Case for Never Throwing Robot Data Away
Flex-π: One Robot Policy, Three Futures, Fifty-Six Ways to Run It
Imagined Motion, Grounded Geometry: How Rest2Art Makes Closed Objects Articulate
ComBodied Agents: When the Human Becomes the World Model
A Sphere Every Robot Hand Can Speak
The Soft Gripper Pareto Frontier: Adaptability Without Surrendering Payload
Dyna-2's Million-Hour Bet: When Human Video Starts Improving Robots
Think Deeply Only When It Matters: DySL-VLA and Action-Aware Dynamic Depth
The 100,000-Hour Bet: How Xiaomi-Robotics-1 Tries to Make Robot Scaling Real
Reason First, Move Second: A Modular Alternative to End-to-End Robot Policies
From BIM to the Build: A Multimodal Agent for Conversational Construction Robotics
The Twin Watches, the QP Dodges: Predictive Safety for a UR10e
One Shot, Twenty Screws: Robot Vision for the Dirty Reality of Appliance Recycling
Better Geometry at the Neural–Symbolic Boundary: DAIoU for Robot Manipulation
REL Hand: A Mechanical Shortcut to More Human-Like Humanoid Dexterity
When the Appliance Becomes the API: Part–Function–State Models for Robot Planning
CISMG-Nav: When a Navigation Agent Stops Forgetting the Building
ReaDy-Go: Teaching Robots to Dodge Photorealistic People in Gaussian-Splat Digital Twins
The Action Is Missing: Turning Human Video into Robot Control
Look Where You Mean: Gaze2Act Gives VLAs a Live Intent Channel
HARP-VLA: Align the Eyes Before the Actions
Make Geometry the Policy: GAM's Bet on 3D-Native Robot Control
π0.7: When a Robot Foundation Model Learns How to Behave
EgoHumanoid: Turning Human First-Person Experience into Humanoid Loco-Manipulation
CycleRL: Teaching a Bicycle to Balance in Simulation—and Trusting It on Asphalt
Feeling the World Without a Force Sensor: UniFP for Legged Loco-Manipulation
The Robot Needs a Memory: Multi-Agent Reasoning Inside a Digital Twin
Language Is Not Just a Prompt: Four Ways Robots Turn Words into Action
Becoming the Path: Planning Motion, Shape, and Stiffness Together
Meaning, Motion, and Two Hands: A Task Graph That Actually Plans
Safety Before Failure: Giving Robot Plans a One-Second Reflex
VL-GRiP3 and the Case for Keeping VLMs Away from the Motor Loop
Think Late, Act Now: TIC-VLA Turns Reasoning Latency into a Control Variable
One Policy, Many Bodies: Inside Qwen-VLA's Generalist Robot Model
VLAbot: A Human-in-the-Loop Control Plane for Long-Horizon Robot Assembly
Teaching Isaac Sim to Feel: A Real-Time Digital Twin for Capacitive Touch
Stop Starting From Noise: Action-to-Action Flow Matching for Fast Robot Control
ACE-Data-0: Turning a Home into a Multimodal Robot-Learning Instrument
Before the Robot Feels It: How N₀-VTLA Predicts Contact to Control It
Metis Gives Foundation Models a Native, Mutable Memory
TurboVLA and the Case for Taking the LLM Out of the Servo Loop
Catching on Optical Expansion: Inside Pixel2Catch's Monocular Sim-to-Real Pipeline
Calibrate the Robot, Randomize the World: PACE for Legged Sim-to-Real
HumanCLAW: The Model Sees the Chair but Loses the Body
Get Up, Adapt, Survive: Recovery Skills for Small Quadrupeds
A Patch Only Needs a Glimpse: Breaking VLA Robots Under Partial Observability
A Thousand Tasks, One Day, and the Case Against Learning Everything End to End
HUG: Turning Everyday Human Grasps Into a Robot-Ready Prior
SkillPlug: Teaching Robot Policies to Reuse the Moves Between Tasks
Expand Before You Compress: TEAM-VLA's Training-Free Shortcut to Faster Robot Policies
One Policy, Three Bodies, Zero Messages: Inside CHORUS
Anchor the Semantics, Align the Actions: A Better Way to Fine-Tune VLAs
Learning the Rule, Not the Gait: Stick-Insect Behavior as a Portable Robot Objective
Two Brains Under Pressure: How UnderwaterVLA Connects Language Models to AUV Control
From Demo to Duty: RL-100's Real-World Reinforcement Learning Flywheel
Distilling a Swarm into 158 Million Parameters: Inside MiniUAV-VLA
Imagining Locomotion: Learning a Neural World Model for Legged Robots
One Model to See, Plan, and Act: Introducing EO-1 for Embodied AI
Video-Action Models for Robot Learning
FastDSAC: Unlocking the Potential of Maximum Entropy RL in High-Dimensional Humanoid Control
A Minimalist Retargeting-Guided Reinforcement Learning Recipe for Dexterous Manipulation
DexNDM: Closing the Reality Gap for Dexterous In-Hand Rotation
Learning Multi-Modal Trajectory Policies for Data-Efficient Robotic Manipulation
ACE: Agentic Control for Embodied Manipulation via Zero-shot Workflow Reasoning
Let the Body Follow: Coupled Egocentric Control for Whole-Body Robot Teleoperation
GigaWorld-Policy-0.5: An Efficient Mixture-of-Transformers Policy for Real-Time Robot Control
Xiaomi-Robotics-1: A Scalable Vision-Language-Action Foundation Model for Mobile Manipulation
Local Policies Enable Zero-shot Long-Horizon Manipulation
Sim-to-Real Reinforcement Learning for Vision-Based Dexterous Manipulation on Humanoids
UniTracker: A Universal Motion Tracking Framework for Humanoid Robots
ViTacFormer: Dexterous Manipulation via Active Vision and High-Resolution Touch Sensing
Masked IRL: LLM-Guided Reward Disambiguation from Demonstrations and Language
FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space
RoboTTT: End-to-End Robotic Assembly with Test-Time Training
Xiaomi-Robotics-U0: A 38B World-Foundation Model for Unified Embodied Perception and Synthesis
SpatialPoint: Spatial-Aware Point Prediction for Embodied Localization
Multi-AUV Scene-Adaptive Embodied Intelligence for Multi-Target Tracking
Speedup Patch: Learning a Plug-and-Play Policy to Accelerate Embodied Manipulation
R3DP: Real-Time 3D-Aware Policy for Embodied Manipulation
VistaVLA: Geometry- and Semantic-Aware 3D Gaussian-Grounded VLA for Robotic Manipulation
SKooP: Symmetric Koopman Predictions for Legged Robot Reinforcement Learning
RobotTT: A Tactile Transformer for Dexterous Manipulation
TAC-LOCO: Unified Whole-Body Control for Contact-Aware Locomotion and Manipulation
TACTIC: Contact-Centric Control for Whole-Arm Manipulation
SIEVE: Structure-Aware Data Selection for Imitation Learning with VLAs