WORLD-MODELSCURRENT2026-04-22

Toward Safe Autonomous Robotic Endovascular Interventions using World Models

Harry Robertshaw, Nikola Fischer, Han-Ru Wu, Andrea Walker Perez, Weiyuan Deng, Benjamin Jackson, Christos Bergeles, Alejandro Granados, Thomas C Booth

This paper demonstrates that world-model-based Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. (TD-MPC2) can autonomously navigate catheters through patient-specific blood vessels with 58% success in Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. and 68% in physical phantom experiments—beating standard Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. by 22 percentage points. It's the first end-to-end validation of autonomous mechanical thrombectomy with Simulation & Sim-to-RealSim-to-real (sim2real)Transferring a policy trained in simulation to a real robot. transfer, showing world models generalize better than model-free Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. to unseen patient anatomies while maintaining safety constraints (Movement, Mechanics & Robot BodyContactPhysical interaction between the robot and an object or surface. forces below rupture threshold).

THE PROBLEM

This paper focuses on world models. This paper demonstrates that world-model-based Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. (TD-MPC2) can autonomously navigate catheters through patient-specific blood vessels with 58% success in Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. and 68% in physical phantom experiments—beating standard Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. by 22 percentage points. It's the first end-to-end validation of autonomous mechanical thrombectomy with Simulation & Sim-to-RealSim-to-real (sim2real)Transferring a policy trained in simulation to a real robot. transfer, showing world models generalize better than model-free Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. to unseen patient anatomies while maintaining safety constraints (Movement, Mechanics & Robot BodyContactPhysical interaction between the robot and an object or surface. forces below rupture threshold). Read the paper by tracking the Core ConceptsTaskThe job the robot is supposed to complete, such as pick-and-place, navigation, or drawer opening. definition, the Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. or data assumptions, and the evidence that supports the claimed improvement.

HOW IT WORKS

1

Task framing

The paper frames the work as world models. Start here because it defines what success means and which assumptions the rest of the method inherits.

2

Core method

This paper demonstrates that world-model-based Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. (TD-MPC2) can autonomously navigate catheters through patient-specific blood vessels with 58% success in Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. and 68% in physical phantom experiments—beating standard Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. by 22 percentage points. It's the first end-to-end validation of autonomous mechanical thrombectomy with Simulation & Sim-to-RealSim-to-real (sim2real)Transferring a policy trained in simulation to a real robot. transfer, showing world models generalize better than model-free Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. to unseen patient anatomies while maintaining safety constraints (Movement, Mechanics & Robot BodyContactPhysical interaction between the robot and an object or surface. forces below rupture threshold). When reading the method section, identify the inputs, the learned or engineered representation, and the Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. or prediction produced by the system.

3

Data and supervision

For robotics work, the data story is part of the method: check whether the system depends on Imitation & Reinforcement LearningTeleoperation (teleop)A human remotely controlling the robot, often to collect demonstrations., Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested., internet video, human labels, or Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. rollouts.

4

Evaluation evidence

The paper should be judged through its Simulation & Sim-to-RealEvaluationMeasuring how well a robot system performs. protocol: what data is used, what Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. or simulator is tested, and which Evaluation & ResearchBaselineA reference method used for comparison. comparisons support the claim. Look for the gap between the headline result and the Simulation & Sim-to-RealDeploymentPutting the trained system on a real robot. setting you would actually care about.

FIGURES

KEY RESULTS

Main contributionConceptual contribution

This paper demonstrates that world-model-based Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. (TD-MPC2) can autonomously navigate catheters through patient-specific blood vessels with 58% success in Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. and 68% in physical phantom experiments—beating standard Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. by 22 percentage points. It's the first end-to-end validation of autonomous mechanical thrombectomy with Simulation & Sim-to-RealSim-to-real (sim2real)Transferring a policy trained in simulation to a real robot. transfer, showing world models generalize better than model-free Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. to unseen patient anatomies while maintaining safety constraints (Movement, Mechanics & Robot BodyContactPhysical interaction between the robot and an object or surface. forces below rupture threshold).

WHY DEVELOPERS SHOULD CARE

This paper demonstrates that world-model-based Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. (TD-MPC2) can autonomously navigate catheters through patient-specific blood vessels with 58% success in Simulation & Sim-to-RealSimulationA virtual environment where robots can be trained or tested. and 68% in physical phantom experiments—beating standard Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. by 22 percentage points. It's the first end-to-end validation of autonomous mechanical thrombectomy with Simulation & Sim-to-RealSim-to-real (sim2real)Transferring a policy trained in simulation to a real robot. transfer, showing world models generalize better than model-free Imitation & Reinforcement LearningReinforcement Learning (RL)Teaching a robot through trial and error using rewards. to unseen patient anatomies while maintaining safety constraints (Movement, Mechanics & Robot BodyContactPhysical interaction between the robot and an object or surface. forces below rupture threshold).

LIMITATIONS

The main limitation to check is whether the claimed behavior holds outside the paper's reported setup. That means testing across different Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. embodiments, scenes, objects, and data distributions.

WHAT COMES NEXT

The practical next step is independent reproduction with clear baselines, ablations, and stress tests. For a developer, the useful follow-up is to map the paper's world models assumptions onto a concrete Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. stack, then test the smallest version of the method that could run end to end.

RELATED PAPERS