Architecture and methods for robotic mobile manipulation system
The present approach relates to integrated planning of robot navigation and manipulator motion in performing tasks. In particular, as discussed herein sensor information, such as visual information, may be used for the robot to locate a target object when it arrives at a working location. Adjustments may be made based on this information that may include moving the robot and planning a manipulator arm motion.
1. A processor-implemented method for manipulating one or more objects, the processor-implemented method comprising acts of:
moving a robot having a robot base and a manipulator coupled to the robot base along a path from an initial location to a task location;
at the task location, moving the manipulator of the robot relative to the task location to acquire images via a camera coupled to the manipulator;
detecting pre-defined interface points of an object, by which the manipulator may grasp the object, within the images acquired by the camera coupled to the manipulator;
determining whether to move the robot base to an updated position at the task location, based upon a position and an orientation of the pre-defined interface points, prior to generating manipulator motion trajectories to manipulate the object with the manipulator, wherein the manipulator is configured to grasp the object at the pre-defined interface points and move the object once grasped as part of a task performed by the robot when the robot is suitably positioned and oriented relative to the object; and
upon determining to move the robot base prior to manipulating the object:
generating robot base motion trajectories;
moving the robot base in accordance with the robot base motion trajectories to reposition the robot base to the updated position at the task location;
generating the manipulator motion trajectories; and
moving the manipulator in accordance with the manipulator motion trajectories to manipulate the object.
2. The processor-implemented method of claim 1 , comprising alerting an operator if no pre-defined interface points are detected.
3. The processor-implemented method of claim 1 , comprising:
upon determining to not move the robot base prior to manipulating the object:
generating the manipulator motion trajectories; and
moving the manipulator in accordance with the manipulator motion trajectories to manipulate the object.
4. The processor-implemented method of claim 1 , wherein a determination of whether to move the robot base prior to manipulating the object is based at least on resolving an energy consumption function.
5. The processor-implemented method of claim 1 , wherein a determination of whether to move the robot base prior to manipulating the object is based at least on reducing or minimizing torque on joints of the manipulator.
6. The processor-implemented method of claim 1 , wherein the manipulator comprises a self-aligning end-effector used to grasp the object at a respective one of the pre-defined interface points.
7. A robotic manipulation system, comprising:
a mobile robot having a robot base, the mobile robot comprising:
one or more sensors configured to generate sensor data of at least a portion of an environment in which the mobile robot is located; and
a first driving system configured to move the robot base of the mobile robot;
a manipulator attached to the robot base, the manipulator comprising:
an effector coupled to the manipulator and configured to interact with objects;
a camera coupled to the manipulator and configured to acquire images of an area accessible by the manipulator; and
a second driving system configured to move the manipulator; and
one or more processing components configured to:
instruct the first driving system to move the robot base of the mobile robot along a path from an initial location to a task location;
move the manipulator relative to the task location to acquire images of the task location via the camera coupled to the manipulator;
detect pre-defined interface points of a respective object within the images with which the manipulator may grasp the respective object;
determine whether to move the robot base to an updated position at the task location, based upon a position and an orientation of the pre-defined interface points, prior to generating manipulator motion trajectories to manipulate the respective object with the manipulator, wherein the manipulator is configured to grasp the respective object at the pre-defined interface points and move the respective object once grasped as part of a task performed by the robot when the robot is suitably positioned and oriented relative to the respective object; and
upon determining to move the mobile robot prior to manipulating the respective object:
generate robot base motion trajectories;
move the robot base of the mobile robot in accordance with the robot base motion trajectories to reposition the robot base;
generate the manipulator motion trajectories; and
move the manipulator in accordance with the manipulator motion trajectories to manipulate the respective object.
8. The robotic manipulation system of claim 7 , wherein the one or more processing components are further configured to alert an operator if no pre-defined interface points are detected.
9. The robotic manipulation system of claim 7 , wherein the one or more processing components are further configured to:
upon determining to not move the robot base prior to manipulating the respective object:
generate the manipulator motion trajectories; and
move the manipulator in accordance with the manipulator motion trajectories to manipulate the respective object.
10. The robotic manipulation system of claim 7 , wherein a determination of whether to move the robot base prior to manipulating the respective object is based at least on resolving an energy consumption function.
11. The robotic manipulation system of claim 7 , wherein a determination of whether to move the robot base prior to manipulating the respective object is based at least on reducing or minimizing torque on joints of the manipulator.
12. The robotic manipulation system of claim 7 , wherein the effector comprises a self-aligning gripper used to grasp the respective object at a respective one of the pre-defined interface points.
13. One or more non-transitory computer-readable media encoding processor-executable routines, wherein the routines, when executed by a processor, cause acts to be performed comprising:
moving a robot having a robot base and a manipulator along a path from an initial location to a task location;
at the task location, moving the manipulator of the robot relative to the task location to acquire images via a camera couple to the manipulator;
detecting pre-defined interface points of an object, within the images, by which the manipulator may grasp the object;
determining whether to move the robot base to an updated position at the task location, based upon a position and an orientation of the pre-defined interface points, prior to generating manipulator motion trajectories to manipulate the object with the manipulator, wherein the manipulator is configured to grasp the object at the pre-defined interface points and move the object once grasped as part of a task performed by the robot when the robot is suitably positioned and oriented relative to the object; and
upon determining to move the robot base prior to manipulating the object:
generating robot base motion trajectories;
moving the robot base in accordance with the robot base motion trajectories to reposition the robot base to the updated position at the task location;
generating the manipulator motion trajectories; and
moving the manipulator in accordance with the manipulator motion trajectories to manipulate the object.
14. The one or more non-transitory computer-readable media of claim 13 , wherein the routines, when executed by the processor, cause further acts to be performed comprising:
upon determining to not move the robot base prior to manipulating the object:
generating the manipulator motion trajectories; and
moving the manipulator in accordance with the manipulator motion trajectories to manipulate the object.
15. The one or more non-transitory computer-readable media of claim 13 , wherein a determination of whether to move the robot base prior to manipulating the object is based at least on resolving an energy consumption function.
16. The one or more non-transitory computer-readable media of claim 13 , wherein a determination of whether to move the robot base prior to manipulating the object is based at least on reducing or minimizing torque on joints of the manipulator.
17. The one or more non-transitory computer-readable media of claim 13 , wherein the routines, when executed by the processor, cause further acts to be performed comprising:
alerting an operator if no pre-defined interface points are detected.