In-Sensor Fall Detection Using Low-Power Finite State Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fall detection solutions in wearable devices are computationally intensive and have high power consumption, making them unsuitable for continuous operation in portable devices like headphones and smart watches.
Innovation Solution
A low power inertial measurement unit (IMU) is used to perform fall detection by integrating a pressure sensor and an application processor, utilizing a finite state machine for shock, altitude change, and steadiness detection without machine learning, allowing the solution to run continuously in low-power devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning techniques are used for fall detection, then detection accuracy is improved, but power consumption increases
Solution Approach 1:
The patent extracts the fall detection functionality from the main application processor and implements it directly in the IMU sensor hardware. This extraction allows the use of simplified detection algorithms (finite state machines) instead of computationally intensive machine learning, significantly reducing power consumption while maintaining adequate detection accuracy for the specific fall detection application
Solution Approach 2:
The patent uses simple, low-cost finite state machine algorithms implemented in hardware rather than complex machine learning models. These simpler algorithms consume minimal power and can run continuously on battery-powered wearable devices without requiring frequent recharging
2Productivity
If fall detection executes on the main application processor, then processing capability is improved, but power consumption increases
Solution Approach 1:
The patent segments the processing functions by separating fall detection processing from the main application processor and dedicating it to the IMU sensor's onboard processing unit. This segmentation allows the main processor to enter low-power states while the IMU continues fall detection with minimal power consumption
Solution Approach 2:
The IMU sensor performs fall detection autonomously using its own onboard processing capabilities and finite state machine implementation. This self-service approach eliminates the need for the high-power main application processor to be actively involved in fall detection, enabling continuous monitoring with minimal energy expenditure
3Reliability
If machine learning techniques are used, then detection reliability is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex machine learning algorithms with simple finite state machine logic that can be implemented directly in hardware. This simplification reduces computational requirements and device complexity while providing sufficient reliability for fall detection through carefully designed state transitions that capture the essential characteristics of fall events
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution provides reliable fall detection with a low false positive rate and minimal power consumption, enabling continuous operation in wearable devices.
Implementation Method 1
The IMU includes one or more motions sensors, such as an accelerometer and a gyroscope
Implementation Method 2
The IMU includes one or more motions sensors, such as an accelerometer and a gyroscope
Implementation Method 3
The IMU receives pressure sensor data from the pressure sensor
Data Source
AI summary
The present disclosure is directed to a device and method for human fall detection solution. Fall detection is performed by a low power inertial measurement unit (IM U) that is communicatively coupled between a pressure sensor and an application processor. The IM U includes one or more motions sensors, such as an accelerometer and gyroscope. The application processor is the main processor of the containing device. The IM U receives pressure sensor data from the pressure sensor, and executes the fall detection using both the pressure sensor data and accelerometer data.