Vision Language Action (VLA) Models for Unmanned Aerial Robotics and Bimanual Manipulation: A Review
Inkyu Sa, Chanoh Park, Hea-Min Lee, Donghee Noh, Ho Seok Ahn
THE PROBLEM
This paper focuses on Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions.. This comprehensive review shows how VLAs—models that combine vision, language, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. in one network—enable robots to perform complex tasks like Manipulation & TasksBimanual manipulationUsing two arms or hands together. and drone Control & PlanningControlThe method used to make the robot move the way you want. from natural language commands and camera images. For developers, it maps 183 papers across Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions. architectures, Robot LearningTrainingThe process of fitting a model using data or experience. methods, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. representations, showing that techniques for coordinating two Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. arms transfer directly to coordinating drone thrust and Movement, Mechanics & Robot BodyGripperA common end-effector used to grasp objects. commands. 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
Task framing
Core method
Data and supervision
Evaluation evidence
KEY RESULTS
This comprehensive review shows how VLAs—models that combine vision, language, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. in one network—enable robots to perform complex tasks like Manipulation & TasksBimanual manipulationUsing two arms or hands together. and drone Control & PlanningControlThe method used to make the robot move the way you want. from natural language commands and camera images. For developers, it maps 183 papers across Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions. architectures, Robot LearningTrainingThe process of fitting a model using data or experience. methods, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. representations, showing that techniques for coordinating two Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. arms transfer directly to coordinating drone thrust and Movement, Mechanics & Robot BodyGripperA common end-effector used to grasp objects. commands.
WHY DEVELOPERS SHOULD CARE
This comprehensive review shows how VLAs—models that combine vision, language, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. in one network—enable robots to perform complex tasks like Manipulation & TasksBimanual manipulationUsing two arms or hands together. and drone Control & PlanningControlThe method used to make the robot move the way you want. from natural language commands and camera images. For developers, it maps 183 papers across Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions. architectures, Robot LearningTrainingThe process of fitting a model using data or experience. methods, and Core ConceptsActionA command the robot sends to its motors, controller, or low-level system. representations, showing that techniques for coordinating two Core ConceptsRobotA physical system with sensors and actuators that can observe the world and take actions. arms transfer directly to coordinating drone thrust and Movement, Mechanics & Robot BodyGripperA common end-effector used to grasp objects. commands.
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 Modern Robot LearningVision-Language-Action model (VLA)A model that takes images and language as input and outputs robot actions. 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.