Motion Direction Estimation Using Bimodal Acceleration Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile devices face challenges in accurately determining motion direction without GPS or external signals, leading to unreliable estimates over time due to growing uncertainties in dead-reckoning applications.
Innovation Solution
A method that uses acceleration data to determine a primary axis of motion by fitting it to a bimodal distribution, selecting the higher peak as the motion direction, and calculating a reliability metric based on peak heights and widths, while also considering orientation data from gyroscopes and magnetometers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If dead-reckoning is used to determine motion direction without GPS or external signals, then the mobile device can operate autonomously in areas without external signals, but the reliability of motion direction estimation decreases significantly over time due to growing uncertainties
Solution Approach 1:
The system continuously monitors motion data from accelerometers and gyroscopes, using feedback loops to detect changes in motion patterns. When external signals become available, the system compares dead-reckoning estimates with actual position data to correct accumulated errors and reset uncertainty levels, thereby maintaining reliability over extended periods
Solution Approach 2:
The system dynamically adjusts processing parameters such as filter coefficients, integration time constants, and sensor weighting factors based on motion characteristics and environmental conditions. This allows optimization of estimation accuracy for different scenarios (walking, running, vehicle transport) and compensates for drift by adapting parameters when external corrections are received
2Reliability
If multiple sensors and complex processing algorithms are used to improve motion direction accuracy, then estimation reliability improves, but device complexity increases
Solution Approach 1:
The system divides motion analysis into distinct phases: acceleration detection, direction determination, and position integration. Each phase uses specialized processing tailored to its specific requirements, allowing optimization of each segment while keeping overall complexity manageable through modular architecture
Solution Approach 2:
The system replaces complex mechanical orientation mechanisms with sensor-based electronic detection using accelerometers and gyroscopes. Software algorithms substitute for complex hardware processing, enabling accurate motion direction determination through computational methods rather than mechanical means
Data Source
AI summary
This disclosure provides devices, computer programs and methods for determining a motion direction. In one aspect, a mobile device includes sensors for measuring acceleration data. The mobile device also includes a processor and a memory that implement a motion direction estimation module configured to determine a primary axis of motion. The motion direction estimation module also determines a motion direction along the primary axis. The determination includes fitting the acceleration data, or data derived therefrom, to a bimodal distribution. A first peak of the bimodal distribution corresponds to a first motion direction along the primary axis, and a second peak corresponds to a second motion direction opposite the first. The motion direction estimation module is configured to estimate the motion direction based on the bimodal distribution. In some implementations, the motion direction estimation module selects the motion direction corresponding to the higher of the peaks as the estimated motion direction.