Controlling a robot to remedy a problem

A non-transitory machine-readable medium storing instructions that, when executed by a processor, cause the processor to implement-a robot intelligence engine that:

• communicates with a highly immersive virtual environment (HIVE) data collector to receive HIVE data characterizing a set of robot tasks executed by a test robot in a HIVE representative of an environment of operation, wherein the robot tasks of the set of robot tasks include at least one robot skill, and wherein the HIVE data is collected by the HIVE data collector from the HIVE;
• stores the set of robot tasks in a database;
• receives sensor data from a problem detecting robot remotely deployed in the environment of operation;
• evaluates the sensor data to determine a cause of a problem detected by the problem detecting robot;
• searches the set of robot tasks to determine if a subset of the robot tasks are potentially employable to remedy the detected problem;
• simulates the subset of robot tasks to determine a likelihood of success for the subset of robot tasks, wherein the simulation generates a set of unsupervised robot tasks that are potentially employable to remedy the detected problem, the set of unsupervised robot tasks are modified versions of the subset of robot tasks;
• selects one of the unsupervised robot tasks to determine a selected robot task for remedying the detected problem;
• selects a robot from a plurality of robots in the environment of operation for executing the selected robot task after the simulation;
• and
• provides the selected robot task to the selected robot operating in the environment of operation to cause the selected robot to execute the selected robot task to attempt to remedy the problem;
• and
• wherein the HIVE data collector:
• sets conditions in the HIVE to emulate conditions of the environment of operation in response to determining that the selected robot task was successfully executed by the selected robot deployed in the environment to remedy the problem based on feedback data, the feedback data characterizing the selected robot task and is provided by the selected robot;
• instructs the test robot in the HIVE to execute the selected robot task;
• and
• tunes and validates the selected robot task based on data collected from the execution by the test robot to convert the selected robot task into a supervised robot task, the supervised robot task being stored in the database;
• and
• providing the supervised robot task to a respective robot of the plurality of robots to cause mechanical features of the respective robot to perform a physical operation of the supervised robot task to remedy a detected problem in the environment of operation.

A method comprising:

• communicating with a highly immersive virtual environment (HIVE) data collector to receive HIVE data characterizing a set of robot tasks executed by a test robot in a HIVE representative of an environment of operation, wherein the robot tasks of the set of robot tasks include at least one robot skill, and wherein the HIVE data is collected by the HIVE data collector from the HIVE;
• storing the set of robot tasks in a database;
• receiving sensor data from a problem detecting robot remotely deployed in the environment of operation;
• evaluating the sensor data to determine a cause of a problem detected by the problem detecting robot;
• searching the set of robot tasks to identify a subset of the robot tasks that are potentially employable to remedy the detected problem;
• simulating the subset of robot tasks to determine a likelihood of success for the subset of robot tasks, wherein the simulating generates a set of unsupervised robot tasks that are potentially employable to remedy the detected problem, the set of unsupervised robot tasks are modified versions of the subset of robot tasks;
• selecting one of the unsupervised robot tasks to determine a selected robot task for remedying the detected problem;
• selecting a robot from a plurality of robots in the environment of operation for executing the selected robot task after the simulation;
• providing the selected robot task to the selected robot operating in the environment of operation to cause the selected robot to execute the selected robot task to attempt to remedy the problem;
• setting, using the HIVE data collector, conditions in the HIVE to emulate conditions of the environment of operation in response to determining that the selected robot task was successfully executed by the selected robot deployed in the environment to remedy the problem based on feedback data, the feedback data characterizing the selected robot task and is provided by the selected robot;
• instructing, using the HIVE data collector, the test robot in the HIVE to execute the selected robot task;
• tuning and validating, using the HIVE data collector, the selected robot task based on data collected from the execution by the test robot to convert the selected robot task into a supervised robot task, the supervised robot task being stored in the database;
• and
• providing the supervised robot task to a respective robot of the plurality of robots to cause mechanical features of the respective robot to perform a physical operation of the supervised robot task to remedy a detected problem in the environment of operation.