Virtual Sensor for Vehicle Dynamics Estimation via Nonlinear Regression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating vehicle dynamics variables like side slip angle, longitudinal and lateral velocities, and tire-road friction coefficient in production vehicles are limited by the use of complex and expensive sensors, and two-step procedures that rely on approximate vehicle models, leading to inaccurate and unbounded estimation errors due to nonlinearities and changing operational conditions.
Innovation Solution
A method using an optimal nonlinear regression function calculated offline from reference data to estimate vehicle dynamics variables, such as side slip angle, by applying the function to real-time measured variables from the Electronic Stability Control system, ensuring estimation accuracy across various operational conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct measurement sensors are used for vehicle dynamics variables, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a virtual sensor that copies the measurement function of complex physical sensors by using a regression-based estimation system. Instead of directly measuring side slip angle, longitudinal velocity, and lateral velocity with expensive sensors, the system computes these variables from readily available ESC system data through offline-optimized regression functions, achieving accurate measurements without physical copying of the measurement hardware
Solution Approach 2:
The patent replaces the mechanical sensor measurement system with a computational estimation system. Rather than using physical sensors to detect vehicle dynamics variables, the invention substitutes a mathematical regression model that processes electronic data from existing vehicle systems (steering angle, wheel velocities, yaw rate, lateral acceleration) to compute the desired variables, eliminating the need for complex mechanical measurement devices
2Device complexity
If two-step estimation methods with approximate models are used, then device complexity is reduced, but measurement precision deteriorates due to model approximations and nonlinearities
Solution Approach 1:
The patent applies preliminary action by performing offline optimization of regression functions before real-time operation. The system pre-computes optimal regression functions using extensive reference data collected under various operating conditions, storing these functions for immediate application during actual vehicle operation. This preliminary preparation eliminates the need for complex real-time model identification and ensures high measurement precision without requiring complex computational resources during critical real-time estimation
Solution Approach 2:
The patent changes the parameter representation from complex nonlinear dynamic models to optimized regression functions with bounded errors. By transforming the estimation problem from solving nonlinear differential equations to applying pre-optimized regression functions, the system maintains measurement precision while dramatically reducing computational complexity. The regression functions are designed with explicit error bounds that account for varying operational conditions without requiring real-time model parameter adjustments
3Measurement precision
If complex nonlinear models are used for accurate estimation, then measurement precision is improved, but productivity decreases due to computational intractability
Solution Approach 1:
The patent copies the essential measurement information from complex nonlinear models through simplified regression functions that preserve accuracy while enabling real-time computation. The regression functions are designed to replicate the behavior of complex models with bounded errors, allowing the system to achieve the same measurement precision as complex models but with computationally tractable operations suitable for real-time embedded systems in production vehicles
Data Source
Figure 1
Figure 2
AI summary
Method for the estimation of at least a variable (β; v x , v y ; ψ, μ) affecting a vehicle dynamics (10), including measuring dynamic variables (MQ) of the vehicle (10) during its motion, calculating in real time an estimate (Formula (I)) of said variable (β; v x , v y ; ψ, μ), on the basis of said measured dynamic variables (MQ), The method includes: calculating (230) said estimate of said at least a variable (β; v x , v y ; ψ, μ) by an estimation procedure (DVS β ; DVS βv ,- DVS βvμ ) comprising taking in account a set of dynamic variables (MQ) measured during the motion of the vehicle (10) over respective time intervals (ny, nw, nψ, nx, nα) and applying on said set of measured dynamic variables (MQ) at least an optimal nonlinear regression function (ƒ* β ; ƒ* x , ƒ* y ; ƒ* β1 , ƒ* β2 , ƒ* ψ 1 , ƒ* ψ 2 ) calculated with respect to said variable (β; v x , v y ; ψ, μ) to estimate to obtain said estimate of said variable (β; v x , v y ; ψ, μ), said optimal non linear regression function (ƒ* β ; ƒ* x , ƒ* y ; ƒ* β1 , ƒ* β2 , ƒ* ψ 1 , ƒ* ψ 2 ) being obtained by an optimal calculation procedure (220) including: on the basis of an acquired set of reference data (Dd) and of said set of dynamic variables (MQ) measured during the motion of the vehicle (10), finding, for a desired accuracy level (ε), a regression function (ƒ* β ; ƒ* x , ƒ* y ; ƒ* β1 , ƒ* β2 , ƒ* ψ 1 , ƒ* ψ 2 ) giving an estimation error lower or equal than said desired accuracy level (ε) in a given set of operative conditions (OC), said acquired set of reference data (Dd) being obtained by acquiring (210) in said given set of operative conditions (OC) a set of reference data (Dd) of variables including variables corresponding to said measured dynamic variables (MQ) of the vehicle (10) and a lateral ( vy) and a longitudinal velocity (vx) of the vehicle (10).