Acceleration Sensor Bias Estimation Using Distance Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calibrating acceleration sensors in vehicle navigation systems face challenges in accuracy due to the influence of vehicle acceleration and posture changes, which affect the measurement accuracy of distance sensors like cameras and LRFs.
Innovation Solution
A system that uses a distance sensor to measure surroundings, calculates vehicle translation and rotation angles, and employs sensor parameter estimation to accurately estimate acceleration sensor bias based on these calculations and sensor measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS-based sensor calibration is used, then acceleration sensor bias can be corrected, but it cannot be used in environments where GPS signals are unavailable or inaccurate
Solution Approach 1:
The patent introduces distance sensors (cameras, LRF) as intermediary devices to measure object distances and calculate vehicle position changes. These sensors serve as mediators between the vehicle and the environment, enabling position estimation without GPS signals. The system uses multiple distance sensors to triangulate vehicle position and calculate translation/rotation angles, providing an alternative calibration method that works in GPS-denied environments.
2Adaptability or versatility
If distance sensors are used to correct acceleration sensor bias, then GPS signal availability is not required, but vehicle acceleration and posture changes affect measurement accuracy
Solution Approach 1:
The patent implements a feedback mechanism where the calculated vehicle translation and rotation angles from distance sensors are continuously compared with acceleration sensor measurements. The system uses this feedback to iteratively refine the acceleration sensor bias estimation, compensating for the effects of vehicle acceleration and posture changes. The calculated angles serve as reference values to correct the acceleration sensor data, improving measurement precision despite dynamic vehicle conditions.
Solution Approach 2:
The patent changes the measurement parameters by using distance sensor data to calculate vehicle translation distance and rotation angles as alternative reference measurements. Instead of relying on GPS position data or raw acceleration sensor output, the system transforms distance sensor measurements into vehicle motion parameters (translation angle, rotation angle) that can be used for bias correction. This parameter transformation enables accurate calibration under varying vehicle conditions.
3Measurement precision
If vehicle posture changes are measured by distance sensors, then position correction can be performed, but posture changes greatly affect measurement accuracy of the distance sensors
Solution Approach 1:
The patent applies preliminary action by calculating and compensating for the effects of vehicle posture changes before using distance sensor measurements for position estimation. The system first estimates vehicle rotation angles from acceleration sensor data, then uses these preliminary estimates to correct the distance sensor measurements. This preliminary compensation removes the harmful effect of posture changes on measurement accuracy, enabling precise position estimation even when the vehicle is moving or changing orientation.
Data Source
AI summary
A driving support device includes a maximum measurement range determination unit adapted to calculate a distance to a plurality of points on a surface of an object present at surroundings of a vehicle. A distance sensor such as a laser range finder or a stereo camera that measures a distance is used in order to estimate bias of an acceleration sensor with high accuracy upon correction of the bias by use of the distance sensor such as the stereo camera or the LRF in a GPS non-receiving section. A sensor parameter estimation unit is adapted to estimate a parameter of the acceleration sensor based on a behavior of an own vehicle obtained from a vehicle translation and rotation angle calculation means adapted to count backwards the vehicle behavior based on distance variation and also based on a measurement value of the acceleration sensor.