Back to results
US 12,517,149 B2Granted
Information processing apparatus, information processing method, and program
AssigneeSony Interactive Entertainment Inc.
InventorYoshinori Ohashi
Life of this patent
Estimating the term…
Abstract
An information processing apparatus 1 sequentially receives a plurality of input data between predetermined time TS and time TE, and uses model information to estimate predetermined estimated data at a time point of time T (TS<T<TE) based on the received input data, the model information having been trained by machine learning so as to estimate the estimated data at the time point of the time T on the basis of the received input data.
Claims20
A system comprising:
- a game controller device;
- an inertial measurement unit included in the game controller device;
- and
- an information processing apparatus comprising processing circuitry configured to perform operations comprising:
- sequentially receiving a plurality of input data generated between predetermined time TS and time TE, the plurality of input data comprising acceleration information output by the inertial measurement unit;
- estimating posture information of the game controller device as a posture quaternion based on the received plurality of input data, wherein each component of the posture quaternion includes a vector representing a direction of a rotation axis represented by a predetermined global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in the acceleration information by inputting the posture quaternion into the first machine learning model, the first machine learning model using the posture quaternion as input data and being in a state of having learned, by machine learning, at least a relation among the posture quaternion, the scale factor error, and the bias error;
- generating calibrated acceleration information using the estimated bias error and scale factor error;
- receiving a sequence of the calibrated acceleration information representing acceleration occurring between the predetermined time TS and time TE;
- using a second machine learning model to estimate a true value of acceleration at a time point T (TS<T<TE) by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the time point T based on the estimated true value of acceleration;
- integrating the velocity value with subsequent acceleration information obtained between the time point T and the time TE to determine a position of the game controller device at the time TE;
- and
- controlling operations of a game application based on the determined position of the game controller device.
11 more elements
A system comprising:
- a game controller device;
- an inertial measurement unit included in the game controller device;
- and
- an information processing apparatus comprising processing circuitry configured to perform operations comprising:
- receiving, from the game controller device, an output of the inertial measurement unit;
- estimating posture information of the game controller device as a posture quaternion on a basis of the output received from the inertial measurement unit, wherein each component of the posture quaternion includes a vector representing a direction of a rotation axis represented by a global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in information regarding acceleration output by the inertial measurement unit included in the game controller device by inputting the posture quaternion into the first machine learning model, the first machine learning model using the posture quaternion as input data and being in a state of having learned, by machine learning, at least a relation between the input data, the scale factor error, and the bias error included in the information regarding the acceleration output by the inertial measurement unit;
- generating calibrated acceleration information using the estimated bias error and scale factor error for a process of calibration of the acceleration output by the inertial measurement unit;
- receiving a sequence of the calibrated acceleration information between a first time point and a second time point;
- using a second machine learning model to estimate a true acceleration value at a third time point between the first time point and the second time point by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the third time point based on the estimated true acceleration value;
- integrating the velocity value with subsequent acceleration information obtained between the third time point and the second time point to determine a position of the game controller device at the second time point;
- and
- controlling operations of a game application based on the determined position of the game controller device.
11 more elements
A system comprising:
- a game controller device;
- an inertial measurement unit included in the game controller device;
- and
- an information processing apparatus comprising processing circuitry configured to perform operations comprising:
- receiving, from the game controller device, an output of the inertial measurement unit;
- estimating posture information of the game controller device on a basis of the output received from the inertial measurement unit, wherein the posture information is represented by a global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in information regarding acceleration output by the inertial measurement unit included in the game controller device by inputting, out of the estimated posture information of the game controller device, information obtained by setting a rotation angle in a gravity direction to a predetermined angle into the first machine learning model, the first machine learning model using, out of the estimated posture information, posture information in which the rotation angle in the gravity direction is set to the predetermined angle as input data and being in a state of having learned, by machine learning, at least a relation between the input data and the bias error included in the information regarding the acceleration output by the inertial measurement unit;
- using the estimated bias error and scale factor error for a process of calibration of the movement acceleration output by the inertial measurement unit to generate calibrated acceleration information;
- receiving a sequence of the calibrated acceleration information between a first time point and a second time point;
- using a second machine learning model to estimate a true acceleration value at a third time point between the first time point and the second time point by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the third time point based on the estimated true acceleration value;
- integrating the velocity value with subsequent acceleration information obtained between the third time point and the second time point to determine a position of the game controller device at the second time point;
- and
- controlling operations of a game application based on the determined position of the game controller device.
11 more elements
A system comprising:
- a game controller device;
- an inertial measurement unit included in the game controller device;
- and
- an information processing apparatus comprising processing circuitry configured to perform operations comprising:
- receiving, from a device including an inertial measurement unit, an output of the inertial measurement unit;
- estimating posture information of the game controller device on a basis of the output received from the inertial measurement unit, wherein the posture information is represented by a global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in information regarding acceleration output by the inertial measurement unit included in the game controller device by inputting, out of the estimated posture information of the game controller device, information excluding a rotation angle in a gravity direction into the first machine learning model, the first machine learning model using, out of the estimated posture information, the information excluding the rotation angle in the gravity direction as input data and being in a state of having learned, by machine learning, at least a relation between the input data and the bias error included in the information regarding the acceleration output by the inertial measurement unit;
- using the estimated bias error and scale factor error for a process of calibration of the acceleration output by the inertial measurement unit to generate calibrated acceleration information;
- receiving a sequence of the calibrated acceleration information between a first time point and a second time point;
- using a second machine learning model to estimate a true acceleration value at a third time point between the first time point and the second time point by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the third time point based on the estimated true acceleration value;
- integrating the velocity value with subsequent acceleration information obtained between the third time point and the second time point to determine a position of the game controller device at the second time point;
- and
- controlling operations of a game application based on the determined position of the game controller device.
11 more elements
A computer-implemented method comprising:
- receiving, from a game controller device including an inertial measurement unit, an output of the inertial measurement unit;
- estimating posture information of the game controller device as a posture quaternion on a basis of the output received from the inertial measurement unit, wherein each component of the posture quaternion includes a vector representing a direction of a rotation axis represented by a global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in information regarding acceleration output by the inertial measurement unit included in the game controller device by inputting the posture quaternion into the first machine learning model, the first machine learning model using the posture quaternion as input data and being in a state of having learned, by machine learning, at least a relation between the input data, the scale factor error, and the bias error included in the information regarding the acceleration output by the inertial measurement unit;
- using the estimated bias error and scale factor error for a process of calibration of the acceleration output by the inertial measurement unit to generate calibrated acceleration information;
- receiving a sequence of the calibrated acceleration information between a first time point and a second time point;
- using a second machine learning model to estimate a true acceleration value at a third time point between the first time point and the second time point by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the third time point based on the estimated true acceleration value;
- integrating the velocity value with subsequent acceleration information obtained between the third time point and the second time point to determine a position of the game controller device at the second time point;
- and
- controlling operations of a game application based on the determined position of the game controller device.
7 more elements
A non-transitory computer-readable storage medium storing computer-readable instructions thereon which, when executed by a computer, cause the computer to perform operations comprising:
- receiving, from a game controller device including an inertial measurement unit, an output of the inertial measurement unit;
- estimating posture information of the game controller device as a posture quaternion on a basis of the output received from the inertial measurement unit, wherein each component of the posture quaternion includes a vector representing a direction of a rotation axis represented by a global coordinate system different from a posture of the game controller device;
- using a first machine learning model to estimate at least a bias error and a scale factor error included in information regarding acceleration output by the inertial measurement unit included in the game controller device by inputting the posture quaternion into the first machine learning model, the first machine learning model using the posture quaternion as input data and being in a state of having learned, by machine learning, at least a relation between the input data, the scale factor error, and the bias error included in the information regarding the acceleration output by the inertial measurement unit;
- using the estimated bias error and scale factor error for a process of calibration of the acceleration output by the inertial measurement unit to generate calibrated acceleration information;
- receiving a sequence of the calibrated acceleration information between a first time point and a second time point;
- using a second machine learning model to estimate a true acceleration value at a third time point between the first time point and the second time point by inputting the sequence of calibrated acceleration information into the second machine learning model;
- determining a velocity value at the third time point based on the estimated true acceleration value;
- integrating the velocity value with subsequent acceleration information obtained between the third time point and the second time point to determine a position of the game controller device at the second time point;
- and
- controlling operations of a game application based on the determined position of the game controller device.
7 more elements
Dependencies are inferred from the claim text.