Patent title: Adaptive Step Detection
Patent number: US 20130191069A1
Patent filing date: 01/18/2013
Patent issue date: 07/25/2013
Time it took for the patent to be issued: Just over 6 months
Inventor: Sourabh Ravindran
Assignee: Texas Instruments Incorporated
U.S. classification: G01C22/006 Pedometers
Number of claims: 7
Today, there are many different types of pedometers that are used by athletes and non-athletes alike. Brands such as Garmin, Fitbit, and Apple make smart watches that allow users to track their steps, distance covered, and floors climbed all while reading text messages and playing music. However, before these complex devices, people still used pedometers to track their steps. Traditional pedometers were worn on clipped to the waist and tracked steps based on the movement of the hips. This patent was filed by Texas Instruments Incorporated for a pedometer that would be worn on the wrist instead of the hip. Devices like this helped pave the way for the popular smart watches worn today.
The main claim of this pedometer is that it can be worn on the wrist and can track steps as accurately as traditional pedometers worn on the hip (figure 1). The another main claim of this device is that it uses three accelerometers to track step data to account for sway and extraneous movements of the arm during daily life. The device also has the capacity to store data, which can be exported to other devices, such as a computer via USB or Bluetooth. In addition, the device has a screen to display step count or distance traveled.
Traditional pedometers were worn on the belt and steps were detected based on the motion of the hips. Movement at the wrist is more complex and can result in more false steps than pedometers worn on the hip. The algorithm used to determine what is registered as a step was altered to account for this more complex motion. To do this, a three axis accelerometer was used to make motion detectable regardless of how the arm was oriented. Data from each axis is filtered and combined by summing the absolute value of each sample. The result is one graph that represents all of the acceleration data in order to get a more accurate depiction of when steps were taken (figure 2).
Using this plot, an adaptive peak detector is utilized in the hardware to quantify the acceleration of each movement. This detector identifies inflection points in the acceleration data collected to identify positive and negative slopes in the accelerations. If the slope regions reach or surpass a threshold value and last for a specified time threshold, then the device registers this as a step. The time restraint helps separate noise from actual step data. The detector then repeats this to track steps over time. Step frequency and the height of the user are determined in order to estimate stride length so that distance covered can also be output to the user. A study conducted showed that this device on the wrist is just as accurate as an older pedometer that was worn on the hip.
Though the mechanisms used to count steps seem rather complex, this device could be used by anyone looking to track their daily steps. This device does not require any difficult training to use so learning how to use the device should not be a limiting factor for this device. Pedometers are used by people of all athletic abilities. If someone wants to begin exercising, this device could be used to track the number of steps accumulated during the day or during a particular workout. An avid runner could use this device to track the distance covered during a run based on stride length and step count. Therefore, this device can be widely used and may be of benefit to anyone trying to increase their physical fitness. Current wrist pedometers have exceeded the functions of this device, incorporating heart rate monitors, swim tracking, GPS tracking, and other technologies. The patent described some of these functions as potential future adaptations/embodiment of this device.
Ravindran, S. (2013). US Patent No. US 2013/0191069A1. Retrieved from https://patents.google.com/patent/US20130191069?oq=intitle%3Aadaptive+intitle%3Astep+intitle%3Adetection