US 12,290,928 B2
Systems and methods for flexible robotic manipulation by fast online load estimation
Yebin Wang, Cambridge, MA (US); Xiaoming Duan, Boston, MA (US); Toshiaki Koike-Akino, Cambridge, MA (US); and Philip Orlik, Cambridge, MA (US)
Assigned to Mitsubishi Electric Research Laboratories, Inc., Cambridge, MA (US)
Filed by Mitsubishi Electric Research Laboratories, Inc., Cambridge, MA (US)
Filed on Mar. 23, 2022, as Appl. No. 17/656,113.
Prior Publication US 2023/0302636 A1, Sep. 28, 2023
Int. Cl. B25J 9/16 (2006.01)
CPC B25J 9/1633 (2013.01) [B25J 9/163 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A manipulator, comprising:
a robotic arm comprising one or more joints configured to move a load according to a task;
a plurality of actuators configured to change a motion of the robotic arm to track a trajectory;
a memory configured to store a model of the manipulator with unknown load information and a mapping function mapping initial poses of the load to corresponding identification trajectories;
an input interface configured to receive an initial pose of the load and the task for moving the load; and
a processor configured to execute stored instructions implementing different modules of the manipulator, including:
a load estimator configured to:
determine using the mapping function, at least one initial configuration of the one or more joints corresponding to the initial pose of the load;
compute an identifiability score of each initial configuration of the at least one initial configuration of the one or more joints;
compare the identifiability score of each initial configuration of the at least one initial configuration with a threshold;
obtain a subset of identification trajectories comprising an identification trajectory for each initial configuration of the at least one initial configuration whose identifiability score exceeds the threshold;
calculate a motion distance between each initial configuration of the at least one initial configuration and a goal configuration for each identification trajectory in the subset of identification trajectories;
determine a trajectory in the subset of identification trajectories for which the calculated motion distance is least among the subset of identification trajectories;
select the determined trajectory as a retrieved identification trajectory in the subset of identification trajectories;
control the plurality of actuators to move the load according to the retrieved identification trajectory, based on the model of the manipulator with unknown load information;
obtain measured motion data corresponding to the movement of the load according to the retrieved identification trajectory and estimated motion data of the load corresponding to motion of the load according to the identification trajectory and estimated based on the model of the manipulator; and
estimate parameters of the load based on the measured motion data of the load and the estimated motion data of the load; and
a performance controller configured to:
obtain a model of the manipulator having the load with the estimated parameters of the load;
determine a performance trajectory to move the load according to the task based on the model of the manipulator having the load with the estimated parameters; and
control the actuators to move the load according to the performance trajectory.