Training robotic manipulators
Methods and systems for training a robotic manipulator. The system may include one or more sensor devices and a robotic manipulator for executing an item grasping strategy to grasp an item. The system may further evaluate the item grasping strategy to determine whether the strategy was successful.
1. A method of training a robotic manipulator, the method comprising:
receiving data regarding an item from a first sensor device;
generating, using a processing device executing instructions stored on a memory to provide a planning module, an item grasping strategy to be executed by the robotic manipulator to obtain an initial grasp on the item, wherein the item grasping strategy is defined by a probability parameter that is related to a number of previous initial grasp attempts;
transmitting, to a robotic manipulator for execution, the item grasping strategy;
receiving data regarding the execution of the item grasping strategy from a second sensor device; and
evaluating the item grasping strategy using the planning module and the data received from at least a second sensor device.
2. The method of claim 1 wherein the received data regarding the item includes at least one of size data, shape data, material data, and weight data.
3. The method of claim 1 further comprising:
altering at least one parameter of the item grasping strategy to provide a corrective item grasping strategy based on the evaluation; and
transmitting, to the robotic manipulator for execution, the corrective grasping strategy.
4. The method of claim 3 wherein the at least one parameter is selected from the group consisting of robotic manipulator position before grasping, pre-grasp manipulations, image processing technique, and response to feedback from the first sensor device.
5. The method of claim 1 wherein the robotic manipulator includes at least one of a robotic hand device, a suction device, and an adhesive strategy for executing the grasping strategy.
6. The method of claim 1 wherein evaluating the item grasping strategy includes determining whether the robotic manipulator grasped the item.
7. The method of claim 1 wherein the first sensor device and the second sensor device are the same device.
8. The method of claim 1 further comprising iterating the steps of:
transmitting, to a robotic manipulator for execution, the item grasping strategy;
receiving data regarding the execution of the item grasping strategy from the second sensor device; and
evaluating the item grasping strategy using the planning module and the data received from the second sensor device.
9. The method of claim 1 further comprising storing data regarding the success rate of the item grasping strategy in a database module.
10. The method of claim 1 further comprising selecting the item to be grasped based on at least one of previous grasp success rates and at least one physical characteristic of the item.
11. The method of claim 1 wherein transmitting the item grasping strategy to the robotic manipulator for execution includes transmitting the item grasping strategy to the robotic manipulator for execution during at least one of a weekend, an overnight period, and a lull period.
12. A system for training a robotic manipulator, the system comprising:
a robotic manipulator for grasping an item;
a first sensor device for gathering data regarding the item;
a second sensor device for gathering data regarding the execution of the item grasping strategy;
a processing device executing instructions stored on a memory to provide a planning module configured to:
generate an item grasping strategy to be executed by the robotic manipulator to obtain an initial grasp on the item, wherein the item grasping strategy is defined by a probability parameter that is related to a number of previous initial grasp attempts,
transmit the item grasping strategy to a robotic manipulator for execution; and
evaluate the item grasping strategy executed by the robotic manipulator using data received from the second sensor device.
13. The system of claim 12 wherein the data regarding the item includes at least one of size data, shape data, material data, and weight data.
14. The system of claim 12 wherein the planning module is further configured to alter at least one parameter of the item grasping strategy based on the evaluation to provide a corrective item grasping strategy, and the robotic manipulator is further configured to execute the corrective item grasping strategy.
15. The system of claim 14 wherein the at least one parameter is selected from the group consisting of robotic manipulator position before grasping, pre-grasp manipulations, image processing technique, and response to feedback from the first sensor device.
16. The system of claim 12 wherein the robotic manipulator includes at least one of a robotic hand device, a suction device, and an adhesive device for executing the grasping strategy.
17. The system of claim 12 wherein evaluating the item grasping strategy includes determining whether the robotic manipulator grasped the item.
18. The system of claim 12 wherein the first sensor device and the second sensor device are the same device.
19. The system of claim 12 wherein the robotic manipulator and the planning module are further configured to iterate the steps of generating the item grasping strategy, transmitting the item grasping strategy, and evaluating the item grasping strategy.
20. The system of claim 12 further comprising a database module for storing data regarding the success rate of the item grasping strategy.
21. The system of claim 12 wherein the item to be grasped is selected based on at least one of previous grasp success rates and at least one physical characteristic of the item.
22. The system of claim 12 wherein the robotic manipulator executes the grasping strategy during at least one of a weekend, an overnight period, and a lull period.