Automatic determination of robot settling states
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for controlling a robot based on settling time of the robot. One of the methods includes executing, by a robot having one or more joints, a motion trajectory to move from a first pose to a second pose; repeatedly computing a settled state confidence score after the robot has begun execution of the motion trajectory; and whenever the current settled state confidence score satisfies a threshold value, executing, by the robot, a command that causes the robot to move.
1 . A method comprising:
executing, by a robot having one or more joints, a motion trajectory to move from a first pose to a second pose;
repeatedly computing a settled state confidence score after the robot has begun execution of the motion trajectory, wherein the settled state confidence score represents a level of confidence that the robot has entered a settled state, and wherein computing the confidence score comprises, at each of multiple time points:
determining, for each of the one or more joints of the robot, a current velocity value based on status data reported by the robot; and
determining, for each of the one or more joints of the robot, a velocity difference between (i) the current velocity value determined from the status data reported by the robot and (ii) a target velocity value specified by the motion trajectory; and
computing, based at least on the current velocity values, and the velocity differences, a current settled state confidence score for the time point; and
whenever the current settled state confidence score satisfies a threshold value, executing, by the robot, a command that causes the robot to move.
2 . The method of claim 1 , wherein the settled state is a state of the robot in which a velocity value of a joint of the robot is within an error band around a predetermined velocity value.
3 . The method of claim 2 , wherein the predetermined velocity value is zero.
4 . The method of claim 1 , wherein computing the confidence score comprises:
obtaining an indicator value that indicates whether the execution of the motion trajectory by the robot is complete; and
computing, based on the current velocity values, the velocity differences, and the indicator value, the current settled state confidence score for the time point.
5 . The method of claim 1 , wherein computing the current settled state confidence score comprises:
computing a variance of the velocity differences over multiple time points up to the current time point.
6 . The method of claim 5 , wherein computing the current settled state confidence score comprises:
determining an update to a previous settled state confidence score based on the variance of the velocity differences and the current velocity values.
7 . The method of claim 4 , wherein the indicator value is a Boolean value.
8 . The method of claim 1 , wherein the multiple time points are separated from each other by a fixed time interval that is dependent on an operating frequency of the robot.
9 . The method of claim 1 , wherein executing, by the robot, the command that causes the robot to move comprises:
performing an action by an end effector of the robot in accordance with the command.
10 . The method of claim 1 , wherein executing, by the robot, the command that causes the robot to move comprises:
executing another motion trajectory to move from the second pose to a third pose in accordance with the command.
11 . A system comprising one or more computers and one or more non-transitory storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
executing, by a robot having one or more joints, a motion trajectory to move from a first pose to a second pose;
repeatedly computing a settled state confidence score after the robot has begun execution of the motion trajectory, wherein the settled state confidence score represents a level of confidence that the robot has entered a settled state, and wherein computing the confidence score comprises, at each of multiple time points:
determining, for each of the one or more joints of the robot, a current velocity value based on status data reported by the robot; and
determining, for each of the one or more joints of the robot, a velocity difference between (i) the current velocity value determined from the status data reported by the robot and (ii) a target velocity value specified by the motion trajectory; and
computing, based at least on the current velocity values, and the velocity differences, a current settled state confidence score for the time point; and
whenever the current settled state confidence score satisfies a threshold value, executing, by the robot, a command that causes the robot to move.
12 . The system of claim 11 , wherein the settled state is a state of the robot in which a velocity value of a joint of the robot is within an error band around a predetermined velocity value.
13 . The system of claim 12 , wherein the predetermined velocity value is zero.
14 . The system of claim 11 , wherein computing the confidence score comprises:
obtaining an indicator value that indicates whether the execution of the motion trajectory by the robot is complete; and
computing, based on the current velocity values, the velocity differences, and the indicator value, the current settled state confidence score for the time point.
15 . The system of claim 11 , wherein computing the current settled state confidence score comprises:
computing a variance of the velocity differences over multiple time points up to the current time point.
16 . The system of claim 15 , wherein computing the current settled state confidence score comprises:
determining an update to a previous settled state confidence score based on the variance of the velocity differences and the current velocity values.
17 . The system of claim 14 , wherein the indicator value is a Boolean value.
18 . The system of claim 11 , wherein the multiple time points are separated from each other by a fixed time interval that is dependent on an operating frequency of the robot.
19 . The system of claim 11 , wherein executing, by the robot, the command that causes the robot to move comprises one of:
performing an action by an end effector of the robot in accordance with the command, or
executing another motion trajectory to move from the second pose to a third pose in accordance with the command.
20 . A non-transitory computer storage medium encoded with instructions that, when executed by one or more computers, cause the one or more computers to perform operations comprising:
executing, by a robot having one or more joints, a motion trajectory to move from a first pose to a second pose;
repeatedly computing a settled state confidence score after the robot has begun execution of the motion trajectory, wherein the settled state confidence score represents a level of confidence that the robot has entered a settled state, and wherein computing the confidence score comprises, at each of multiple time points:
determining, for each of the one or more joints of the robot, a current velocity value based on status data reported by the robot; and
determining, for each of the one or more joints of the robot, a velocity difference between (i) the current velocity value determined from the status data reported by the robot and (ii) a target velocity value specified by the motion trajectory; and
computing, based at least on the current velocity values, and the velocity differences, a current settled state confidence score for the time point; and
whenever the current settled state confidence score satisfies a threshold value, executing, by the robot, a command that causes the robot to move.