Machine-learned robot fleet management for value chain networks
A system includes a fleet resources data store that maintains a fleet resource inventory indicating fleet resources that can be assigned to perform tasks. For each fleet resource, the inventory indicates features of each fleet resource and a respective status. A set of task definitions is accessible to an intelligence layer to facilitate improving task definition based on feedback from task-specific outcomes. The system receives a job request for a robotic fleet to perform a job and determines a job definition data structure indicating a set of tasks to be performed for the job. The system applies an outcome of performing a task by a resource assigned to perform the task to a machine learning system of the intelligence layer that facilitates improving, based on the outcome, the set of task definitions. The system updates the set of task definitions based on a result of applying the machine learning system.
1. A system, comprising:
a computer-readable storage system that stores:
a fleet resources data store that maintains a fleet resource inventory that indicates:
a plurality of fleet resources that can be assigned to perform at least one task in a set of tasks, and
for each respective fleet resource of the plurality of fleet resources, a set of features of the respective fleet resource and a status of the respective fleet resource; and
a set of task definitions that is accessible to an intelligence layer to facilitate improving task definition based on feedback from task-specific outcomes; and
a set of processors that executes a set of computer-readable instructions, wherein the set of processors collectively:
receives a job request for a robotic fleet to perform a job, wherein:
the robotic fleet includes a set of robot operating units,
at least one robot of the set of robot operating units includes a control interface module and a physical interface module, and
the control interface module controls the physical interface module;
determines a job definition data structure based on the job request, wherein the job definition data structure indicates a set of tasks that are to be performed in performance of the job;
determines a robotic fleet configuration data structure corresponding to the job based on the set of tasks and the fleet resource inventory, wherein the robotic fleet configuration data structure assigns a plurality of resources selected from the fleet resource inventory to the set of tasks indicated in the job definition data structure based on the set of task definitions, the respective set of features of each fleet resource, and the respective status of each fleet resource;
deploys the robotic fleet to perform the job, wherein performing the job includes performing at least one task by at least one resource of the plurality of resources assigned to perform the at least one task;
applies an outcome of performing the at least one task by at least one resource of the plurality of resources assigned to perform the at least one task to a machine learning system of the intelligence layer that facilitates improving, based on the outcome, the set of task definitions; and
updates the set of task definitions based on a result of applying the outcome to the machine learning system,
wherein:
the set of task definitions indicates a type of end effector of at least one robot operating unit of the set of robot operating units,
updating the set of task definitions includes communicating with the control interface module to update the type of end effector of the at least one robot operating unit of the set of robot operating units, and
updating the type of end effector includes controlling the physical interface module, by the control interface module, to update the type of end effector of the at least one robot operating unit of the set of robot operating units.
2. The system of claim 1 , wherein the outcome of performing the at least one task includes failure data associated with performing the at least one task.
3. The system of claim 1 , wherein the outcome of performing the at least one task is extracted from a job completion data set.
4. The system of claim 1 , wherein the outcome of performing the at least one task is determined by a workflow simulation system that simulates performing the at least one task.
5. The system of claim 1 , wherein applying the outcome of performing the at least one task to the machine learning system includes applying the outcome to a deep learning system for optimizing a task definition or task assignment across fleet resources.
6. The system of claim 1 , wherein an assignment of a plurality of resources selected from the fleet resource inventory to the set of tasks defined in the job definition data structure is further based on at least one of an environment of the job, a budget for the job, or a timeline for completion of the job.
7. The system of claim 1 , wherein the set of task definitions is updated for an instance of a fleet configuration data structure corresponding to the job.
8. The system of claim 1 , wherein the set of processors collectively updates a second set of task definitions for a second job based on a result of applying the outcome of performing the at least one task to the machine learning system.
9. The system of claim 1 , wherein the set of processors executes the set of computer-readable instructions cooperatively with at least one of a fleet configuration system, a fleet resource scheduling system, a fleet security system, or a fleet utilization system.
10. The system of claim 1 , wherein the machine learning system utilizes at least one neural network to facilitate improving the set of task definitions based on a set of features, including features extracted from the job request.
11. A method of assigning robotic fleet resources to tasks, comprising:
storing, in a computer-readable storage system, a fleet resource inventory that indicates a plurality of fleet resources that can be assigned to perform at least one task in a set of tasks, and for each respective fleet resource of the plurality of fleet resources, a set of features of the respective fleet resource and a status of the respective fleet resource;
storing, in the computer-readable storage system, a set of task definitions that is accessible to an intelligence layer to facilitate improving task definition based on feedback from task-specific outcomes;
receiving a request for a robotic fleet to perform a job, wherein:
the robotic fleet includes a set of robot operating units,
at least one robot of the set of robot operating units includes a control interface module and a physical interface module, and
the control interface module controls the physical interface module;
determining a job definition data structure based on the request, wherein the job definition data structure indicates a set of tasks that are to be performed in performance of the job;
determining a robotic fleet configuration data structure corresponding to the job based on the set of tasks and a fleet resource inventory, and for each respective fleet resource of the plurality of fleet resources, the fleet resource inventory indicates a set of features of the respective fleet resource, a set of configuration requirements of the respective fleet resource, and a status of the respective fleet resource, wherein the robotic fleet configuration data structure assigns a plurality of resources selected from the fleet resource inventory to the set of tasks defined in the job definition data structure based on the set of task definitions, the respective set of features of each fleet resource, and the respective status of each fleet resource;
deploying the robotic fleet to perform the job, wherein performing the job includes performing at least one task by at least one resource of the plurality of resources assigned to perform the at least one task;
applying an outcome of performing the at least one task by at least one resource of the plurality of resources assigned to perform the at least one task to a machine learning system of the intelligence layer that facilitates improving, based on the outcome, the set of task definitions; and
updating the set of task definitions based on a result of applying the outcome to the machine learning system, wherein:
the set of task definitions indicates a type of end effector of at least one robot operating unit of the set of robot operating units,
updating the set of task definitions includes communicating with the control interface module to update the type of end effector of the at least one robot operating unit of the set of robot operating units, and
updating the type of end effector includes controlling the physical interface module, by the control interface module, to update the type of end effector of the at least one robot operating unit of the set of robot operating units.
12. The method of claim 11 , wherein the outcome of performing the at least one task includes failure data associated with performing the at least one task.
13. The method of claim 11 , wherein the outcome of performing the at least one task is extracted from a job completion data set.
14. The method of claim 11 , wherein the outcome of performing the at least one task is determined by a workflow simulation system that simulates performing the at least one task.
15. The method of claim 11 , wherein applying the outcome of performing the at least one task to the machine learning system includes applying the outcome to a quantum optimization system for optimizing task definition or task assignment across fleet resources.
16. The method of claim 11 , wherein an assignment of a plurality of resources selected from the fleet resource inventory to the set of tasks defined in the job definition data structure is further based on at least one of an environment of the job, a budget for the job, or a timeline for completion of the job.
17. The method of claim 11 , wherein the set of task definitions is updated for an instance of a fleet configuration data structure corresponding to the job.
18. The method of claim 11 , further including updating a second set of task definitions for a second job request based on a result of applying the outcome of performing the at least one task to the machine learning system.
19. The method of claim 11 , further including operating cooperatively with at least one of a fleet configuration system, a fleet resource scheduling system, a fleet security system, or a fleet utilization system.
20. The method of claim 11 , wherein the machine learning system utilizes at least one neural network to facilitate improving the set of task definitions based on a set of features, including features extracted from a job request.