18379597. SIMULATING MULTIPLE ROBOTS IN VIRTUAL ENVIRONMENTS simplified abstract (GOOGLE LLC)

From WikiPatents
Jump to navigation Jump to search

SIMULATING MULTIPLE ROBOTS IN VIRTUAL ENVIRONMENTS

Organization Name

GOOGLE LLC

Inventor(s)

Matthew Bennice of San Jose CA (US)

Paul Bechard of Ogdensburg NY (US)

SIMULATING MULTIPLE ROBOTS IN VIRTUAL ENVIRONMENTS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18379597 titled 'SIMULATING MULTIPLE ROBOTS IN VIRTUAL ENVIRONMENTS

Simplified Explanation

Implementations are provided for coupling multiple robot controllers to a single virtual environment in order to generate training examples for machine learning models. In these implementations, a simulated virtual environment is created with an interactive object and multiple robot avatars that are controlled independently by external robot controllers. Sensor data from each robot avatar is provided to its corresponding robot controller, and joint commands for actuating the robot avatars' joints are received from the controllers. The robot avatars' joints are then actuated accordingly, allowing them to interact with the interactive object in the virtual environment.

  • The patent describes a method for connecting multiple robot controllers to a virtual environment.
  • The virtual environment includes an interactive object and multiple robot avatars.
  • The robot avatars are controlled independently and simultaneously by external robot controllers.
  • Sensor data from each robot avatar is sent to its corresponding controller.
  • Joint commands for actuating the robot avatars' joints are received from the controllers.
  • The robot avatars' joints are actuated based on the received joint commands.
  • This allows the robot avatars to interact with the interactive object in the virtual environment.

Potential Applications

  • Training machine learning models: The system can generate training examples for machine learning models by simulating interactions between multiple robot avatars and an interactive object in a virtual environment.

Problems Solved

  • Limited availability of physical robot avatars: By using virtual robot avatars, the system allows for the simultaneous control and interaction of multiple robots without the need for physical robots.

Benefits

  • Scalability: The system can easily scale up the number of robot avatars and controllers in the virtual environment, allowing for large-scale training and testing of machine learning models.
  • Cost-effective: By using virtual robot avatars instead of physical robots, the system reduces the cost associated with purchasing and maintaining multiple robots.
  • Safety: The use of virtual environments eliminates the risk of physical harm or damage that may occur during robot interactions in the real world.


Original Abstract Submitted

Implementations are provided for operably coupling multiple robot controllers to a single virtual environment, e.g., to generate training examples for training machine learning model(s). In various implementations, a virtual environment may be simulated that includes an interactive object and a plurality of robot avatars that are controlled independently and contemporaneously by a corresponding plurality of robot controllers that are external from the virtual environment. Sensor data generated from a perspective of each robot avatar of the plurality of robot avatars may be provided to a corresponding robot controller. Joint commands that cause actuation of one or more joints of each robot avatar may be received from the corresponding robot controller. Joint(s) of each robot avatar may be actuated pursuant to corresponding joint commands. The actuating may cause two or more of the robot avatars to act upon the interactive object in the virtual environment.