Sensor Fusion for Machine State Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional systems for determining machine state, such as those used in heavy equipment, face inaccuracies in position estimation during GPS signal unavailability, multipath errors, or when signals are unreliable, due to the lack of accuracy checking in GPS signals.
Innovation Solution
A method and system utilizing Inertial Measurement Units (IMUs) with Kalman filter modules to fuse acceleration and angular rate measurements, generating refined estimates of joint angles, and solving kinematic equations to determine real-time position, velocity, and acceleration of machine components, even in the absence of reliable GPS signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS signals are used to determine machine position, then position information can be obtained, but accuracy deteriorates during GPS signal unavailability, multipath errors, or unreliable signals
Solution Approach 1:
The patent combines multiple sensing systems (GPS receiver, IMU, odometer, and vision-based localization) into a unified sensor fusion system. The GPS receiver provides position data when available, the IMU provides acceleration and angular rate data for dead reckoning, the odometer provides wheel rotation data, and the vision system provides map-matching localization. These diverse sensors are merged through Kalman filtering to produce a reliable position estimate that maintains accuracy even when individual sensors fail or provide poor data quality.
Solution Approach 2:
The patent introduces an intermediary evaluation mechanism that assesses the quality and reliability of GPS signals before using them for position estimation. The system evaluates GPS signal characteristics and determines when GPS data is unreliable due to multipath errors, signal blockage, or excessive noise. When GPS quality falls below thresholds, the system automatically transitions to alternative localization methods (IMU dead reckoning, map-matching) rather than directly using degraded GPS data, thus maintaining position accuracy.
2Adaptability or versatility
If conventional GPS-based positioning is used, then position can be determined, but the system fails to provide accurate estimates during dead-reckoning periods when GPS signals are unavailable
Solution Approach 1:
The patent implements a dynamic positioning system that automatically adapts its methodology based on real-time conditions. The system continuously monitors GPS signal availability and quality, and dynamically switches between different positioning modes: GPS-based positioning when signals are strong and available, IMU-based dead reckoning when GPS is unavailable, and map-matching localization when GPS quality degrades. This dynamic adaptation ensures the system maintains position estimation accuracy across varying operational conditions without manual intervention.
Solution Approach 2:
The patent changes the operational parameters of the positioning system based on environmental conditions. When GPS signals are unavailable or unreliable, the system transitions from GPS-dependent parameters to IMU and odometer parameters for dead reckoning. The Kalman filter dynamically adjusts its process noise covariance and measurement noise covariance parameters based on the selected positioning mode, optimizing the fusion weights for each sensor according to current operational conditions and maintaining accurate position estimates.
3Ease of operation
If GPS signals are used without accuracy checking, then position information is obtained, but the system cannot detect multipath errors, position jumps, or signal reliability issues
Solution Approach 1:
The patent incorporates feedback mechanisms that continuously monitor the quality and consistency of position estimates from multiple sensors. The system evaluates GPS signal characteristics, compares position estimates across different sensor sources, and detects anomalies such as multipath errors, position jumps, or unrealistic velocity changes. When inconsistencies are detected, the feedback loop triggers re-evaluation of sensor data quality, switches to alternative positioning methods, or flags the position estimate as unreliable, providing both simplicity of automatic operation and robust reliability verification.
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
This approach provides accurate and reliable real-time machine state determination, enhancing operational precision and safety by improving position estimation accuracy and reducing reliance on GPS signals.
Implementation Method 1
receiving, with at least one processor, from each of a plurality of Inertial Measurement Units (IMU's) mounted on different components of the machine, a time series of signals indicative of acceleration and angular rate of motion measurements
Implementation Method 2
fusing the signals received from each of the IMU's on a separate component of the machine with a separate Kalman filter module of the at least one processor
Data Source
AI summary
A method of determining the real time state of a machine includes receiving acceleration and angular rate of motion measurements from IMU's mounted on components of a machine. Fusing signals received from the IMU's with separate Kalman filter modules by combining an acceleration measurement and an angular rate of motion measurement from each IMU to estimate an output joint angle for the component on which the IMU is mounted. Estimated and measured values of the output joint angle for each component are combined, a kinematic equation is solved to determine a real time value for at least one of position, velocity, and acceleration of the component at successive timesteps, and the determined real time values are applied to control movement of each component.