Back to results
US 12,642,453 B2Granted
Calculation device, calculation method, and program recording medium
AssigneeNEC Corporation
InventorsChenhui Huang, Kenichiro Fukushi, Zhenwei Wang
Life of this patent
Estimating the term…
Abstract
In order to calculate, with a high degree of accuracy, step lengths for both feet on the basis of a physical quantity relating to the movement of one foot, this calculation device includes: a detection unit that generates a walking waveform using sensor data and detects a walking event from the generated walking waveform, the sensor data being based on a physical quantity that is measured by a sensor positioned on one foot of a walker and relates to the movement of the foot; and a step length calculation unit that calculates the step lengths for both left and right feet on the basis of the timing of the detected walking event.
Claims7
A calculation device comprising:
- one or more memories storing instructions;
- and
- one or more processors configured to execute the instructions to:
- receive sensor data via wireless communication from a data acquisition device installed in footwear of a pedestrian, the data acquisition device including an acceleration sensor and an angular velocity sensor;
- convert the received sensor data from a local coordinate system set in the data acquisition device to a world coordinate system set with respect to a ground;
- generate time-series data associated with walking of the pedestrian in the world coordinate system using the converted sensor data;
- calculate a spatial angle including a plantar angle using at least one of spatial acceleration and spatial angular velocity in the time-series data, and generate time-series data of the plantar angle;
- detect, in a gait waveform of the plantar angle for two gait cycles, a time at which the gait waveform becomes minimum and a time at which the gait waveform becomes maximum;
- calculate a midpoint time between the detected minimum time and the detected maximum time as a start point of one gait cycle, and calculate a next midpoint time as an end point of the one gait cycle;
- cut out, from the time-series data, a gait waveform for the one gait cycle from the start point to the end point;
- generate a gait waveform using the sensor data based on a physical quantity related to movement of a foot of the pedestrian measured by the data acquisition device;
- detect a gait event from the generated gait waveform;
- calculate step lengths of left and right feet based on a timing of the detected gait event;
- calculate asymmetry of the step lengths of the left and right feet;
- calculate a value obtained by dividing the difference between the step lengths of the left and right feet by a sum of the step lengths of the left and right feet as the asymmetry of the step lengths of the left and right feet;
- estimate a physical condition of the pedestrian based on the calculated asymmetry of the step lengths of the left and right feet using a machine learning model;
- and
- transmit information relating to the estimated physical condition, to encourage the pedestrian to make a decision, to a display device via communication network for displaying the information on a screen of the display device.
15 more elements
A calculation method for causing a computer to execute:
- receiving sensor data via wireless communication from a data acquisition device installed in footwear of a pedestrian, the data acquisition device including an acceleration sensor and an angular velocity sensor;
- converting the received sensor data from a local coordinate system set in the data acquisition device to a world coordinate system set with respect to a ground;
- generating time-series data associated with walking of the pedestrian in the world coordinate system using the converted sensor data;
- calculating a spatial angle including a plantar angle using at least one of spatial acceleration and spatial angular velocity in the time-series data, and generate time-series data of the plantar angle;
- detecting, in a gait waveform of the plantar angle for two gait cycles, a time at which the gait waveform becomes minimum and a time at which the gait waveform becomes maximum;
- calculating a midpoint time between the detected minimum time and the detected maximum time as a start point of one gait cycle, and calculate a next midpoint time as an end point of the one gait cycle;
- cutting out, from the time-series data, a gait waveform for the one gait cycle from the start point to the end point;
- generating a gait waveform using the sensor data based on a physical quantity related to movement of a foot of the pedestrian measured by the data acquisition device;
- detecting a gait event from the generated gait waveform;
- calculating step lengths of left and right feet based on a timing of the detected gait event;
- calculating asymmetry of the step lengths of the left and right feet;
- calculating a value obtained by dividing the difference between the step lengths of the left and right feet by a sum of the step lengths of the left and right feet as the asymmetry of the step lengths of the left and right feet;
- estimating a physical condition of the pedestrian based on the calculated asymmetry of the step lengths of the left and right feet using a machine learning model;
- and
- transmitting information relating to the estimated physical condition, to encourage the pedestrian to make a decision, to a display device via communication network for displaying the information on a screen of the display device.
12 more elements
A non-transitory program recording medium recorded with a program for causing a computer to execute:
- processing of sensor data received via wireless communication from a data acquisition device installed in footwear of a pedestrian, the data acquisition device including an acceleration sensor and an angular velocity sensor;
- processing of converting the received sensor data from a local coordinate system set in the data acquisition device to a world coordinate system set with respect to a ground;
- processing of generating time-series data associated with walking of the pedestrian in the world coordinate system using the converted sensor data;
- processing of calculating a spatial angle including a plantar angle using at least one of spatial acceleration and spatial angular velocity in the time-series data, and generate time-series data of the plantar angle;
- processing of detecting, in a gait waveform of the plantar angle for two gait cycles, a time at which the gait waveform becomes minimum and a time at which the gait waveform becomes maximum;
- processing of calculating a midpoint time between the detected minimum time and the detected maximum time as a start point of one gait cycle, and calculate a next midpoint time as an end point of the one gait cycle;
- processing of cutting out, from the time-series data, a gait waveform for the one gait cycle from the start point to the end point;
- processing of generating a gait waveform using the sensor data based on a physical quantity related to movement of a foot of the pedestrian measured by the data acquisition device;
- processing of detecting a gait event from the generated gait waveform;
- and
- processing of calculating step lengths of left and right feet based on a timing of the detected gait event;
- and
- processing of calculating asymmetry of the step lengths of the left and right feet;
- processing of calculating a value obtained by dividing the difference between the step lengths of the left and right feet by a sum of the step lengths of the left and right feet as the asymmetry of the step lengths of the left and right feet;
- processing of estimating a physical condition of the pedestrian based on the calculated asymmetry of the step lengths of the left and right feet using a machine learning model;
- and
- processing of transmitting information relating to the estimated physical condition, to encourage the pedestrian to make a decision, to a display device via communication network for displaying the information on a screen of the display device.
14 more elements
Dependencies are inferred from the claim text.