Power assist system and storage medium
A power assist system according to the present disclosure includes: a robot that is worn by a user, the robot including a drive source that assists action of the user; an interference force estimation unit that estimates an interference force given from the user to the robot; an admittance model that generates a target velocity in a virtual object having a predetermined dynamic characteristic, the target velocity corresponding to the interference force estimated by the interference force estimation unit; and a control unit, in which the interference force estimation unit includes: a disturbance observer that detects a disturbance based on the momentum of the robot; and an estimation unit that estimates the interference force depending on the disturbance detected by the disturbance observer, using an ESN learning model for which learning about the interference force depending on the disturbance is performed while weights in a reservoir layer are stochastically assigned.
1 . A power assist system comprising:
a robot that is worn by a user, the robot including a drive source that assists action of the user;
an interference force estimation unit that estimates an interference force given from the user to the robot;
an admittance model that generates a target velocity in a virtual object having a predetermined dynamic characteristic, the target velocity corresponding to the interference force estimated by the interference force estimation unit; and
a control unit that drives the drive source by a control torque that allows a velocity of the robot to follow the target velocity, wherein:
the interference force estimation unit includes:
a disturbance observer that detects a disturbance based on a momentum of the robot; and
an estimation unit that estimates the interference force depending on the disturbance detected by the disturbance observer, using an (echo state network) ESN learning model for which learning about the interference force depending on the disturbance is performed while weights in a reservoir layer are stochastically assigned.
2 . The power assist system according to claim 1 , wherein:
the ESN learning model is configured to perform learning relevant to the interference force depending on the disturbance, based on teaching data for the interference force, when an operation mode is a learning mode; and
the estimation unit estimates the interference force depending on the disturbance detected by the disturbance observer, using the ESN learning model, when the operation mode is an estimation mode.
3 . The power assist system according to claim 1 , wherein:
the ESN learning model is configured to perform learning using a difference between the interference force and the disturbance as teaching data, when an operation mode is a learning mode; and
the estimation unit estimates a value resulting from adding the interference force estimated using the ESN learning model and the disturbance detected by the disturbance observer, as the interference force, when the operation mode is an estimation mode.
4 . The power assist system according to claim 1 , wherein:
the disturbance observer is a momentum-based observer (MBO).
5 . A non-transitory storage medium storing a control program causing a computer to execute an assist process by a power assist system including a robot that is worn by a user, the robot including a drive source that assists action of the user, wherein:
the control program being configured to cause the computer to execute:
a process of estimating an interference force given from the user to the robot;
a process of generating a target velocity in a virtual object having a predetermined dynamic characteristic, the target velocity corresponding to the interference force; and
a process of driving the drive source by a control torque that allows a velocity of the robot to follow the target velocity;
the control program being configured to cause the computer to, in the process of estimating the interference force, execute:
a process of detecting a disturbance based on a momentum of the robot by a disturbance observer; and
a process of estimating the interference force depending on the disturbance detected by the disturbance observer, using an ESN learning model for which learning about the interference force depending on the disturbance is performed while weights in a reservoir layer are stochastically assigned.