Training and operating industrial robots
Robots may manipulate objects based on sensor input about the objects and/or the environment in conjunction with data structures representing primitive tasks and, in some embodiments, objects and/or locations associated therewith. The data structures may be created by instantiating respective prototypes during training by a human trainer.
1. A robot for manipulating objects within an environment based on physical perception thereof, the robot comprising:
at least one appendage for manipulating an object;
a controller for operating the at least one appendage;
a perception system for receiving physical information about the object and the environment from at least one sensor;
an instance database comprising a plurality of instances at least some of which specify tasks to be performed by the appendage on the object, each task specifying an object and an action to be performed thereon and being defined in generic terms without specifying spatial coordinates; and
a task-execution module, responsive to the perception system, for causing the robot to execute the tasks via commands to the controller, wherein the task-execution module generates, on the fly based on the physical information, a trajectory corresponding to the task in a coordinate system associated with the robot.
2. The robot of claim 1 , wherein the task-execution module continuously monitors the object and, based thereon, causes the robot appendage to move toward the object without computing trajectory coordinates.
3. The robot of claim 2 , wherein the task-execution module causes movement of the robot appendage toward the object to continue as long as a pre-condition associated with the movement is satisfied, as determined by the task execution module based on the monitored object.
4. The robot of claim 3 , wherein the pre-condition comprises absence of physical contact between the appendage and the object.
5. The robot of claim 1 , wherein the perception system includes at least one camera for acquiring images of the object.
6. The robot of claim 5 , wherein the perception system further includes a computer vision system.
7. The robot of claim 1 , wherein the perception system includes at least one touch-sensitive detector.