Simulating multiple robots in virtual environments
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.
1. A method implemented using one or more processors, comprising:
simulating a three-dimensional virtual environment that includes 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;
operating the plurality of robot avatars at frequencies that correspond to real-world frequencies of the respective robot controllers that control the plurality of robot avatars;
receiving, from each robot controller of the plurality of robot controllers, joint commands that cause actuation of one or more joints of the respective robot avatar of the plurality of robot avatars that is controlled by the robot controller;
actuating one or more joints of each robot avatar of the plurality of robot avatars pursuant to corresponding joint commands;
to account for a deviation of a lagging robot avatar of the plurality of robot avatars from an expected operational frequency of the lagging robot avatar, temporarily pausing operations of one or more other robot avatars of the plurality of robot avatars; and
generating and storing training data for a machine learning model based on interactions between the plurality of robot avatars and the virtual environment.
2. The method of claim 1 , further comprising operating a simulated world clock of the virtual environment at a given frequency, wherein the frequencies at which the plurality of robot avatars are operated are less than or equal to the given frequency of the simulated world clock.
3. The method of claim 2 , wherein the pausing comprises pausing the simulated world clock until the lagging robot avatar advances to a next stage of operation.
4. The method of claim 3 , further comprising pausing operations of one or more other robot avatars of the plurality of robot avatars until the lagging robot avatar advances to the next stage of operation.
5. The method of claim 1 , further comprising generating, for each robot avatar of the plurality of robot avatars, a directed acyclic graph with nodes that represent components of the robot avatar.
6. The method of claim 5 , wherein the directed acyclic graph is a dependency graph in which edges between nodes represent dependencies between the components represented by the nodes.
7. The method of claim 6 , wherein at least one node representing a simulated sensor of the robot avatar imposes a delay on output of the simulated sensor being passed up the directed acyclic graph, wherein the delay corresponds to a frequency of a real-world sensor corresponding to the simulated sensor.
8. The method of claim 6 , wherein one or more nodes of the directed acyclic graph represent a simulated environmental condition of the virtual environment.
9. The method of claim 6 , wherein one or more nodes of the directed acyclic graph represent a simulated condition of a simulated sensor of the robot avatar.
10. A system comprising one or more processors and memory storing instructions that, in response to execution by the one or more processors, cause the one or more processors to:
simulate a three-dimensional virtual environment that includes 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;
operate the plurality of robot avatars at frequencies that correspond to real-world frequencies of the respective robot controllers that control the plurality of robot avatars;
receive, from each robot controller of the plurality of robot controllers, joint commands that cause actuation of one or more joints of the respective robot avatar of the plurality of robot avatars that is controlled by the robot controller;
actuate one or more joints of each robot avatar of the plurality of robot avatars pursuant to corresponding joint commands;
to account for a deviation of a lagging robot avatar of the plurality of robot avatars from an expected operational frequency of the lagging robot avatar, temporarily pause operations of one or more other robot avatars of the plurality of robot avatars; and
generate and store training data for a machine learning model based on interactions between the plurality of robot avatars and the virtual environment.
11. The system of claim 10 , further comprising instructions to operate a simulated world clock of the virtual environment at a given frequency, wherein the frequencies at which the plurality of robot avatars are operated are less than or equal to the given frequency of the simulated world clock.
12. The system of claim 11 , wherein the instructions to pause include instructions to pause the simulated world clock until the lagging robot avatar advances to a next stage of operation.
13. The system of claim 12 , further comprising instructions to pause operations of one or more other robot avatars of the plurality of robot avatars until the lagging robot avatar advances to the next stage of operation.
14. The system of claim 10 , further comprising instructions to generate, for each robot avatar of the plurality of robot avatars, a directed acyclic graph with nodes that represent components of the robot avatar.
15. The system of claim 14 , wherein the directed acyclic graph is a dependency graph in which edges between nodes represent dependencies between the components represented by the nodes.
16. The system of claim 15 , wherein at least one node representing a simulated sensor of the robot avatar imposes a delay on output of the simulated sensor being passed up the directed acyclic graph, wherein the delay corresponds to a frequency of a real-world sensor corresponding to the simulated sensor.
17. The system of claim 15 , wherein one or more nodes of the directed acyclic graph represent a simulated environmental condition of the virtual environment.
18. The system of claim 15 , wherein one or more nodes of the directed acyclic graph represent a simulated condition of a simulated sensor of the robot avatar.
19. At least one non-transitory computer-readable medium comprising instructions that, in response to execution by one or more processors, cause the one or more processors to:
simulate a three-dimensional virtual environment that includes 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;
operate the plurality of robot avatars at frequencies that correspond to real-world frequencies of the respective robot controllers that control the plurality of robot avatars;
receive, from each robot controller of the plurality of robot controllers, joint commands that cause actuation of one or more joints of the respective robot avatar of the plurality of robot avatars that is controlled by the robot controller;
actuate one or more joints of each robot avatar of the plurality of robot avatars pursuant to corresponding joint commands;
to account for a deviation of a lagging robot avatar of the plurality of robot avatars from an expected operational frequency of the lagging robot avatar, temporarily pause operations of one or more other robot avatars of the plurality of robot avatars; and
generate and store training data for a machine learning model based on interactions between the plurality of robot avatars and the virtual environment.
20. The at least one non-transitory computer-readable medium of claim 19 , further comprising instructions to operate a simulated world clock of the virtual environment at a given frequency, wherein the frequencies at which the plurality of robot avatars are operated are less than or equal to the given frequency of the simulated world clock.