RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
The paper describes vision-language-action models that represent robot actions alongside visual and language inputs.
Machines, companies, technologies, industries, applications, components, people, research, rules, glossary terms and attributed news, ranked in one list.
73 results for “Cameras”
Also matched: vision, computer vision, imaging, perception, sensorThe paper describes vision-language-action models that represent robot actions alongside visual and language inputs.
Computer vision uses computational methods to extract information from images and video.
Sensor fusion combines measurements from multiple sensors to estimate a system or environment more reliably than a single stream alone.
Vision 60 is a modular quadruped uncrewed ground vehicle built for mobility across rough terrain, with perception-assisted walking and route-level autonomy for…
DJI develops consumer, enterprise and agricultural uncrewed aircraft and imaging systems.
Dieter Fox studies robotics, perception and state estimation; the institutional page is the authority for current roles.
Fei-Fei Li is a Stanford professor whose work includes computer vision and human-centered artificial intelligence.
R1 is a compact Unitree humanoid available in Air, standard and EDU configurations. Its product page emphasizes agile motion, multimodal voice-and-image interaction and…
Kodiak Driver is an autonomous-driving system integrated with Class 8 trucks for long-haul freight, using a modular sensor and compute architecture designed for fleet…
Remotely operated deep-sea research vehicle aboard Falkor (too), equipped for imaging, sampling, and scientific instrument deployment.
Vision-guided tractor implement that mechanically hoes within crop rows around individual plants.
Ouster develops digital lidar sensors and perception software for industrial and mobility uses.
Figure 03 is Figure's current third-generation humanoid platform, designed around its Helix AI system. Specifications and deployment results below are attributed to…
ID. Buzz AD is a pre-series autonomous passenger shuttle built by Volkswagen Commercial Vehicles with Mobileye driving technology for MOIA's turnkey mobility-service…
Digital laparoscopic platform offering surgeon-controlled robotic instrumentation and visualization features for authorized procedures.
Computational methods that extract information from images or video, including detection, tracking, segmentation and depth estimation.
Optimus is Tesla's general-purpose humanoid development program. Public performance and production statements should be treated as company claims until independently…
X10 is an autonomous enterprise quadcopter for inspection, public safety and related operations.
Aurora Driver is an automated-driving system developed initially for highway freight operations.
Waymo Driver is the hardware-and-software automated-driving system used in Waymo's ride-hailing operations. Service availability is geographically limited and changes…
LaserWeeder is a tractor-towed agricultural system that uses cameras, computer vision and lasers for targeted weed control.
Sensors on drones, vehicles or field robots collect imagery and measurements to assess crop conditions.
OS1 is a mid-range digital lidar family for robotics, industrial and mobility perception.
Mobile crop-harvesting robot developed to identify and pick ripe fruit in commercial production environments.
Walker C1 is a commercial service humanoid positioned for customer-facing interaction, guided service and adaptable movement in public indoor spaces.
Walker S2 is an industrial humanoid designed for continuous factory work. UBTECH documents autonomous battery swapping, binocular stereo vision and load handling across…
AKINCI is a twin-engine high-altitude uncrewed combat aircraft designed for long-endurance surveillance, command support and multi-payload defense missions.
CAMCOPTER S-100 is an uncrewed helicopter system for land- and ship-based missions that carries interchangeable sensing payloads without requiring a prepared runway.
Falco Xplorer is a remotely piloted fixed-wing aircraft developed for persistent strategic surveillance with automatic takeoff and landing and an open payload…
MQ-4C Triton is a high-altitude long-endurance uncrewed aircraft designed to provide persistent maritime intelligence, surveillance and reconnaissance over large ocean…
Raven B is a lightweight, hand-launched small uncrewed aircraft system intended to give small units portable daytime and infrared reconnaissance.
Skeldar V-200 is a medium-range remotely piloted helicopter for land and maritime sensing, controlled through high-level autonomous commands from a remote pilot station.
Motional's IONIQ 5 Robotaxi is an all-electric passenger vehicle integrated with the company's SAE Level 4 autonomous-driving system and multi-modal sensor suite.
SuperDrive is PlusAI's Level 4 virtual-driver system for factory-built autonomous trucks being developed with commercial-vehicle manufacturers.
WeRide Robobus is a purpose-built Level 4 electric shuttle without conventional driver controls, intended for public transport, campuses and fixed-route connector…
Robosweeper S6 is a Level 4 road-cleaning vehicle for municipal sanitation fleets, combining autonomous driving with sweeping and water-based cleaning equipment.
GXR is WeRide's purpose-built Level 4 robotaxi platform, designed around a spacious passenger cabin, drive-by-wire controls and redundant autonomous-driving hardware.
Modular portable autonomous underwater vehicle used for survey, monitoring, and search applications.
Portable observation-class remotely operated vehicle with configurable thrusters, sensors, and manipulators.
Experimental rotorcraft that demonstrated powered, controlled flight on Mars and later performed scouting flights.
Stationary Mars lander that used a robotic arm to place geophysical instruments on the surface.
Small hopping lunar excursion vehicle deployed during the SLIM landing mission.
Mars Exploration Rover that conducted long-duration surface science in Meridiani Planum.
Stationary Mars lander equipped with a robotic arm to excavate and analyze soil and ice in the northern plains.
Six-wheeled lunar rover carried by Chandrayaan-3's Vikram lander for in-situ elemental measurements near the lunar south polar region.
Small rover delivered by Mars Pathfinder and the first wheeled vehicle to operate on another planet.
Small transformable lunar robot deployed by SLIM to move on the surface and photograph the lander.
Mars Exploration Rover that investigated rocks, soils, and geological evidence at Gusev Crater.
Camera- and machine-learning-equipped spraying system that identifies crops and weeds for targeted application.
A sensor assembly that measures acceleration and angular motion, often using accelerometers and gyroscopes.
The work studies closed-loop grasping learned from a large collection of robot interaction data.
A sensing method that transmits radio waves and analyzes reflections to estimate properties such as range, direction or relative velocity.
The problem of estimating a system's location while building or updating a map of an initially uncertain environment.
Mobile robots and drones carry cameras or other sensors to observe equipment, structures and sites.
Response robots can help search unstable, remote or hazardous areas while carrying cameras and sensors.
Autonomous tractor using cameras and machine learning to perform bounded tillage work under remote monitoring.
Portable tethered remotely operated vehicle designed for underwater inspection with an articulating camera head and optional tools.
Single-port robotic-assisted surgical system that routes an articulating camera and instruments through one cannula for indicated procedures.
Camera-guided tractor implement designed to apply treatments selectively to individual plants.
Anduril develops defense autonomy software, sensors and uncrewed air and maritime systems.
Carbon Robotics develops computer-vision-guided agricultural equipment for weed control.
Figure develops general-purpose humanoid robots and the Helix vision-language-action system.
Autonomous navigation combines perception, localization, planning and control so a machine can move toward goals with limited direct human input.
Embodied AI studies intelligent agents that perceive and act through a body in a physical or simulated environment.
The E-Model is Starship's sidewalk delivery robot for short-distance food, grocery and package trips, using a locked cargo bay and sensor-rich autonomous navigation.
Walker X is a bipedal service humanoid with vision-based navigation, hand-eye coordination and human-interaction functions.
The Waabi Driver is an autonomous-trucking system combining AI software, sensors and compute for integration into factory-built heavy trucks.
Vision-guided tractor implement designed to cultivate around crop plants within and between rows.
Ground robots, drones and fixed sensors can support monitoring and incident assessment under human oversight.
Figure described Helix 02 as a vision-language-action system for whole-body control of Figure 03 and published demonstrations of long-horizon manipulation. Results are…
A method of piloting or viewing a drone using a live camera feed from the aircraft's perspective; operational rules still apply.
A drone operation in which the remote pilot cannot continuously see the aircraft with unaided vision as required for visual-line-of-sight operation; authorization…
A set of methods that fit model behavior from data for tasks such as prediction, classification, perception or control.