Inertial Vehicle Positioning Using Mixture-Model Sideslip Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile object positioning systems face accuracy issues when the steering angle changes momentarily, leading to decreased estimation of attitude angles.
Innovation Solution
A mobile object positioning device that utilizes a sensor information obtainment unit, sideslip angle estimation unit, and inertial positioning unit to estimate sideslip angles based on a mixture model integrating multiple motion models, correcting sensor values and performing inertial positioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a predefined table or steady-state assumption is used to estimate the sideslip angle, then the device complexity is reduced, but the measurement precision of the attitude angle decreases when the steering angle changes momentarily
Solution Approach 1:
The patent transitions from static estimation methods (predefined tables, steady-state assumptions) to a dynamic estimation approach using a mixture model that continuously adapts to changing vehicle states. The mixture model dynamically weights multiple motion models (steady-state cornering model and transient response model) based on real-time steering angle changes, allowing accurate sideslip angle estimation during both steady-state and transient conditions without increasing device complexity.
Solution Approach 2:
The patent changes the estimation parameters by introducing a mixture model that adjusts weighting factors based on steering angle rate of change. When steering angle changes slowly, the steady-state model is weighted higher; when steering angle changes rapidly, the transient response model is weighted higher. This parameter adaptation resolves the contradiction by maintaining measurement precision across different driving conditions without requiring complex additional sensors or devices.
2Measurement precision
If multiple motion models are integrated using a mixture model to estimate the sideslip angle, then the measurement precision of the attitude angle is improved, but the device complexity increases
Solution Approach 1:
The patent segments the sideslip angle estimation problem into two distinct motion models: a steady-state cornering model for normal driving conditions and a transient response model for steering changes. Each model is optimized for specific conditions, and the mixture model selectively combines them based on the current driving state. This segmentation approach improves measurement precision by using the most appropriate model for each condition while keeping the overall system manageable through clear division of functions.
Solution Approach 2:
The mixture model acts as an intermediary that bridges the steady-state and transient response models. It calculates weighting factors based on steering angle rate of change and combines the outputs of both motion models to produce the final sideslip angle estimation. This intermediary approach allows the system to leverage the strengths of both simple models (low complexity) while achieving accurate estimation across all driving conditions (high precision) without requiring a single complex model.
Data Source
AI summary
The present disclosure has an object of providing a mobile object positioning device positioning a mobile object using a sensor with high accuracy. The mobile object positioning device according to the present disclosure includes: a sensor information obtainment unit obtaining a sensor value on a mobile object, the sensor value being detected by a sensor; a sideslip angle estimation unit estimating a sideslip angle of the mobile object using the sensor value; and an inertial positioning unit performing inertial positioning of the mobile object using the sensor value and the sideslip angle, wherein the sideslip angle estimation unit estimates the sideslip angle based on the sensor value and a mixture model obtained by weighting a plurality of motion models on the mobile object based on state quantities of the mobile object and integrating the plurality of motion models.