This New Robotics System Can Easily Learn a New Task

An artificial intelligence robotics system was designed to learn a new task just by seeing it done once. The non-profit artificial intelligence research company, OpenAI, has trained the robotics system only from simulation and applied it on a physical robot. The company’s mission is to build safe artificial general intelligence and ensure that its benefits are openly distributed.
Robotic system that can learn new tasks
[Image Source: OpenAI]

The robotics system

A vision network and an imitation network power the system, which allows the robot to learn a new behavior that was demonstrated by a simulator. The robotic system can then replicate that specific behavior in different setups within the real world. A built-in camera within the robot captures an image that is ingested by the vision network and subsequently outputs state representing an object’s position. Following the research group’s previous work, the vision network is trained with hundreds of thousands of virtual images that has various perturbations of lighting, textures, and objects. However, the vision network doesn’t practice or train using real images.
The imitation network works exactly as its name suggest. By observing a demonstration, the network processes visualized movements then imitate the task’s intent from a starting configuration. This allows the network to round up the demonstration to a new setting.
OpenAI's vision and imitation network
[Image Source: OpenAI]
For each task, thousands of demonstrations are used to train the imitation network. A training session consists of two demonstrations that do exactly the same task. The first demonstration is given to the network which then has to observe the second demonstration only once. Supervised learning is put to place to predict the demonstrator’s action at that observation. The robot is able to predict actions effectively by learning how to infer the main substance of the task using the first demonstration.

By the team’s continuous work, the imitation network received a new algorithm called one-shot imitation learning. This algorithm permits a human to dish out a new task by performing it in Virtual Reality. From a single demonstration, the robotics system can solve the same task by starting from a random configuration.

Block stacking task

Equipped with the vision and imitation networks, the robot was put to test by getting it to perform a block stacking routine demonstrated via virtual reality that was controlled by a human. The team supplied the robot with training data consisting of pairs of trajectories that stack blocks into a corresponding set of towers in the same order but commencing from a different start state.

The team had to introduce some amount of noise into the outputs of the system’s scripted policy in order for the imitation network to learn a robust policy. Through the use of noise data, the robotics system was able to demonstrate how to re-group itself when something goes wrong. This resulted in a more intelligent imitation network that tackles the anomalies and disturbances from an imperfect script policy.

If you’re interested in being part of this robotics project, you can join OpenAI in their headquarters in San Francisco.
Previous
Next Post »
My photo

Hi, I`m Sostenes, Electrical Technician and PLC`S Programmer.
Everyday I`m exploring the world of Electrical to find better solution for Automation. I believe everyday can become a Electrician with the right learning materials.
My goal with BLOG is to help you learn Electrical.
Related Posts Plugin for WordPress, Blogger...

Label

KITAIFA NEWS KIMATAIFA MICHEZO BURUDANI SIASA TECHNICAL ARTICLES f HAPA KAZI TU. LEKULE TV EDITORIALS ARTICLES DC DIGITAL ROBOTICS SEMICONDUCTORS MAKALA GENERATOR GALLERY AC EXPERIMENTS MANUFACTURING-ENGINEERING MAGAZETI REFERENCE IOT FUNDAMENTAL OF ELECTRICITY ELECTRONICS ELECTRICAL ENGINEER MEASUREMENT VIDEO ZANZIBAR YETU TRANSDUCER & SENSOR MITINDO ARDUINO RENEWABLE ENERGY AUTOMOBILE SYNCHRONOUS GENERATOR ELECTRICAL DISTRIBUTION CABLES DIGITAL ELECTRONICS AUTOMOTIVE PROTECTION SOLAR TEARDOWN DIODE AND CIRCUITS BASIC ELECTRICAL ELECTRONICS MOTOR SWITCHES CIRCUIT BREAKERS MICROCONTROLLER CIRCUITS THEORY PANEL BUILDING ELECTRONICS DEVICES MIRACLES SWITCHGEAR ANALOG MOBILE DEVICES CAMERA TECHNOLOGY GENERATION WEARABLES BATTERIES COMMUNICATION FREE CIRCUITS INDUSTRIAL AUTOMATION SPECIAL MACHINES ELECTRICAL SAFETY ENERGY EFFIDIENCY-BUILDING DRONE NUCLEAR ENERGY CONTROL SYSTEM FILTER`S SMATRPHONE BIOGAS POWER TANZIA BELT CONVEYOR MATERIAL HANDLING RELAY ELECTRICAL INSTRUMENTS PLC`S TRANSFORMER AC CIRCUITS CIRCUIT SCHEMATIC SYMBOLS DDISCRETE SEMICONDUCTOR CIRCUITS WIND POWER C.B DEVICES DC CIRCUITS DIODES AND RECTIFIERS FUSE SPECIAL TRANSFORMER THERMAL POWER PLANT cartoon CELL CHEMISTRY EARTHING SYSTEM ELECTRIC LAMP ENERGY SOURCE FUNDAMENTAL OF ELECTRICITY 2 BIPOLAR JUNCTION TRANSISTOR 555 TIMER CIRCUITS AUTOCAD C PROGRAMMING HYDRO POWER LOGIC GATES OPERATIONAL AMPLIFIER`S SOLID-STATE DEVICE THEORRY DEFECE & MILITARY FLUORESCENT LAMP HOME AUTOMATION INDUSTRIAL ROBOTICS ANDROID COMPUTER ELECTRICAL DRIVES GROUNDING SYSTEM BLUETOOTH CALCULUS REFERENCE DC METERING CIRCUITS DC NETWORK ANALYSIS ELECTRICAL SAFETY TIPS ELECTRICIAN SCHOOL ELECTRON TUBES FUNDAMENTAL OF ELECTRICITY 1 INDUCTION MACHINES INSULATIONS ALGEBRA REFERENCE HMI[Human Interface Machines] INDUCTION MOTOR KARNAUGH MAPPING USEUL EQUIATIONS AND CONVERSION FACTOR ANALOG INTEGRATED CIRCUITS BASIC CONCEPTS AND TEST EQUIPMENTS DIGITAL COMMUNICATION DIGITAL-ANALOG CONVERSION ELECTRICAL SOFTWARE GAS TURBINE ILLUMINATION OHM`S LAW POWER ELECTRONICS THYRISTOR USB AUDIO BOOLEAN ALGEBRA DIGITAL INTEGRATED CIRCUITS FUNDAMENTAL OF ELECTRICITY 3 PHYSICS OF CONDUCTORS AND INSULATORS SPECIAL MOTOR STEAM POWER PLANTS TESTING TRANSMISION LINE C-BISCUIT CAPACITORS COMBINATION LOGIC FUNCTION COMPLEX NUMBERS ELECTRICAL LAWS HMI[HUMANI INTERFACE MACHINES INVERTER LADDER DIAGRAM MULTIVIBRATORS RC AND L/R TIME CONSTANTS SCADA SERIES AND PARALLEL CIRCUITS USING THE SPICE CIRCUIT SIMULATION PROGRAM AMPLIFIERS AND ACTIVE DEVICES BASIC CONCEPTS OF ELECTRICITY CONDUCTOR AND INSULATORS TABLES CONDUITS FITTING AND SUPPORTS CONTROL MOTION ELECTRICAL INSTRUMENTATION SIGNALS ELECTRICAL TOOLS INDUCTORS LiDAR MAGNETISM AND ELECTROMAGNETISM PLYPHASE AC CIRCUITS RECLOSER SAFE LIVING WITH GAS AND LPG SAFETY CLOTHING STEPPER MOTOR SYNCHRONOUS MOTOR AC METRING CIRCUITS APPS & SOFTWARE BASIC AC THEORY BECOME AN ELECTRICIAN BINARY ARITHMETIC BUSHING DIGITAL STORAGE MEMROY ELECTRICIAN JOBS HEAT ENGINES HOME THEATER INPECTIONS LIGHT SABER MOSFET NUMERATION SYSTEM POWER FACTORS REACTANCE AND IMPEDANCE INDUCTIVE RESONANCE SCIENTIFIC NOTATION AND METRIC PREFIXES SULFURIC ACID TROUBLESHOOTING TROUBLESHOOTING-THEORY & PRACTICE 12C BUS APPLE BATTERIES AND POWER SYSTEMS ELECTROMECHANICAL RELAYS ENERGY EFFICIENCY-LIGHT INDUSTRIAL SAFETY EQUIPMENTS MEGGER MXED-FREQUENCY AC SIGNALS PRINCIPLE OF DIGITAL COMPUTING QUESTIONS REACTANCE AND IMPEDANCE-CAPATIVE RECTIFIER AND CONVERTERS SEQUENTIAL CIRCUITS SERRIES-PARALLEL COMBINATION CIRCUITS SHIFT REGISTERS BUILDING SERVICES COMPRESSOR CRANES DC MOTOR DRIVES DIVIDER CIRCUIT AND KIRCHHOFF`S LAW ELECTRICAL DISTRIBUTION EQUIPMENTS 1 ELECTRICAL DISTRIBUTION EQUIPMENTS B ELECTRICAL TOOL KIT ELECTRICIAN JOB DESCRIPTION LAPTOP THERMOCOUPLE TRIGONOMENTRY REFERENCE UART WIRELESS BIOMASS CONTACTOR ELECTRIC ILLUMINATION ELECTRICAL SAFETY TRAINING FILTER DESIGN HARDWARE INDUSTRIAL DRIVES JUNCTION FIELD-EFFECT TRANSISTORS NASA NUCLEAR POWER SCIENCE VALVE WWE oscilloscope 3D TECHNOLOGIES COLOR CODES ELECTRIC TRACTION FEATURED FLEXIBLE ELECTRONICS FLUKE GEARMOTORS INTRODUCTION LASSER MATERIAL PID PUMP SEAL ELECTRICIAN CAREER ELECTRICITY SUPPLY AND DISTRIBUTION MUSIC NEUTRAL PERIODIC TABLES OF THE ELEMENTS POLYPHASE AC CIRCUITS PROJECTS REATORS SATELLITE STAR DELTA VIBRATION WATERPROOF