| 主权项 |
1. A method, comprising:
receiving first sensor data from at least one first sensor, wherein the first sensor is a digital camera oriented to view an arena; determining a target position from the first sensor data, said target position comprising at least one of a position in space and an orientation for a gripper in association with a robot arm; starting to issue at least one first instruction to at least one of the robot arm and the gripper in order to move the gripper to the target position; receiving feedback sensor data from at least one force feedback sensor, associated with at least one of the robot arm and the gripper, to determine a failure in carrying out the at least one first instruction; receiving second sensor data from the at least one first sensor; determining a successful gripping of an object from the second sensor data; receiving verification sensor data from at least one second sensor, in response to the determining of the successful gripping; issuing at least one second instruction to the robot arm in order to move the arm to a predetermined position to release the grip of the gripper; registering the successful gripping and the first sensor data to a first computer program stored in a non-transitory computer readable medium; registering an unsuccessful gripping and the first sensor data to the first computer program, in response to a determined failure in carrying out the at least one first instruction; predicting a likelihood for the successful gripping of a next object from new first sensor data using the information registered in the first computer program; and issuing at least one new first instruction to at least one of the robot arm and the gripper, to grip the next object, based on the predicted likelihood. |
