KR-20260064310-A - APPARATUS AND METHOD FOR ESTIMATING VEHICLE BEHAVIOR

KR20260064310AKR 20260064310 AKR20260064310 AKR 20260064310AKR-20260064310-A

Abstract

The present invention relates to a vehicle behavior determination device and a method thereof. A vehicle behavior determination device according to an embodiment of the present invention may include a sensor for acquiring dynamic characteristics of a vehicle and a processor electrically connected to the sensor. The processor may determine an estimated lateral velocity by fusing a first estimated lateral velocity acquired based on the dynamic characteristics of the vehicle in a first physical model and a second estimated lateral velocity acquired based on a first kinematic model using a first weight, determine an estimated roll angle by fusing a first estimated roll angle acquired based on the estimated lateral velocity in a second kinematic model and a second estimated roll angle acquired based on a second physical model using a second weight, and determine an estimated pitch angle by fusing a first estimated pitch angle acquired based on the estimated lateral velocity in a second kinematic model and a second estimated pitch angle acquired based on a second physical model using a third weight.

Inventors

정래욱
강성욱
최승용
최기훈
유승한
조완기

Assignees

현대자동차주식회사
기아 주식회사
한국기술교육대학교 산학협력단

Dates

Publication Date: 20260507
Application Date: 20241031

Claims (15)

A sensor for acquiring the dynamic characteristics of a vehicle; and It includes a processor electrically connected to the above sensor, and The above processor The estimated lateral velocity is determined by fusing the first estimated lateral velocity obtained based on the dynamic characteristics of the vehicle based on the first physical model and the second estimated lateral velocity obtained based on the first kinematic model using the first weight, and The estimated roll angle is determined by fusing the first estimated roll angle obtained based on the estimated lateral velocity in the second kinematic model and the second estimated roll angle obtained based on the second physical model using a second weight, and A vehicle behavior determination device characterized by determining an estimated pitch angle by fusing a first estimated pitch angle obtained based on the estimated lateral velocity in the second kinematic model and a second estimated pitch angle obtained based on the second physical model using a third weight.
In Article 1, The above processor Determining the derivative value of the above second estimated lateral velocity, and Determining the lateral velocity error between the first estimated lateral velocity and the estimated lateral velocity, and The lateral velocity error is determined by reflecting the first weighting factor to the lateral velocity error, and A vehicle behavior determination device characterized by determining the lateral velocity by integrating the result of summing the derivative of the second estimated lateral velocity and the lateral velocity error.
In Article 2, The above processor A vehicle behavior determination device characterized by determining the derivative value of the second estimated lateral velocity based on the dynamic characteristics of the vehicle, the estimated roll angle, and the estimated pitch angle.
In Article 1, The above processor A vehicle behavior determination device characterized by determining the first estimated roll angle and the first estimated pitch angle based on a kinematic relationship between acceleration and angle.
In Article 4, The above processor A vehicle behavior determination device characterized by determining the second estimated roll angle and the second estimated pitch angle based on a suspension tension-compression approximation model.
In Article 5, The above processor The first estimated roll angle and the second estimated roll angle are fused using a first complementary filter that uses the second weight as the cutoff frequency, and A vehicle behavior determination device characterized by fusing the first estimated pitch angle and the second estimated pitch angle using a second complementary filter that uses the third weight as a cutoff frequency.
In Article 1, The first weight, the second weight, and the third weight are A vehicle behavior determination device characterized by using artificial intelligence learned to reduce a first error between the estimated lateral velocity and the measured lateral velocity, a second error between the estimated roll angle and the measured roll angle, and a third error between the estimated pitch angle and the measured pitch angle.
In Article 7, The first weight, the second weight, and the third weight are A vehicle behavior determination device characterized in that each of the above first error, the above second error, and the above third error is set to reduce the magnitude of the mean squared error between the measured values and the estimated values obtained at a plurality of timings.
A step of determining an estimated lateral velocity by fusing a first lateral velocity obtained based on the dynamic characteristics of a vehicle acquired by a sensor based on a first physical model and a second lateral velocity obtained based on a first kinematic model; A step of determining an estimated roll angle by fusing a first roll angle obtained based on the estimated lateral velocity based on a second kinematic model and a second roll angle obtained based on a second physical model; and A step of determining an estimated pitch angle by fusing a first pitch angle obtained based on the estimated lateral velocity based on the second kinematic model and a second pitch angle obtained based on the second physical model; A method for determining the behavior of a vehicle including
In Article 9, The step of determining the above estimated lateral velocity is A step of determining the derivative value of the second estimated lateral velocity; A step of determining the lateral velocity error between the first estimated lateral velocity and the estimated lateral velocity; A step of determining a lateral velocity error by reflecting the first weighting factor to the lateral velocity error; and A method for determining the behavior of a vehicle, characterized by including a step of determining the estimated lateral velocity by integrating the result of summing the derivative of the second estimated lateral velocity and the lateral velocity error.
In Article 10, The step of determining the derivative value of the second estimated lateral velocity above A method for determining the behavior of a vehicle, characterized by determining the derivative value of the second estimated lateral velocity based on the dynamic characteristics of the vehicle, the estimated roll angle, and the estimated pitch angle.
A step of determining the estimated lateral velocity by fusing the first estimated lateral velocity and the second estimated lateral velocity using the first weight; A step of determining the estimated roll angle by fusing the first estimated roll angle and the second estimated roll angle using a second weight; A step of determining an estimated pitch angle by fusing a first estimated pitch angle and a second estimated pitch angle using a third weight; and A step of learning the first weight to reduce the first error between the estimated lateral velocity and the measured lateral velocity; A method for determining the behavior of a vehicle including
In Article 12, The step of learning the first weight above A step of determining multiple first errors; A method for determining the behavior of a vehicle characterized by learning the first weight to reduce the magnitude of the mean squared error of the plurality of first errors.
In Article 12, A step of determining a second error between the estimated roll angle and the measured roll angle; A step of determining a third error between the estimated pitch angle and the measured pitch angle; and A method for determining the behavior of a vehicle, characterized by further including a step of learning the second weight and the third weight to reduce the magnitude of the second error and the third error.
In Article 13, The step of learning the first weight, the second weight, and the third weight A method for determining the behavior of a vehicle, characterized by being performed to reduce the magnitude of a loss function including the sum of the first error, the second error, and the third error.

Description

Apparatus and Method for Estimating Vehicle Behavior The present invention relates to a device and method for estimating the behavior of a vehicle, and more specifically, to a technology for estimating the operation of a vehicle that can increase reliability. Technology development for Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicles is actively underway to enhance driving convenience and passenger safety. For driver assistance systems and autonomous driving, it is necessary to accurately determine the behavior of a vehicle. Using sensors mounted on the vehicle to determine vehicle behavior is common and can be the most accurate method. However, some sensors are very expensive, so they are not suitable for installation in mass-produced vehicles. For example, sensors for directly measuring lateral velocity, roll angle, and pitch angle are generally not installed in mass-produced vehicles. In cases where sensors for acquiring lateral velocity, roll angle, and pitch angle are not installed, a method for estimating lateral velocity, roll angle, and pitch angle using a pre-designed dynamic model can be used. There may be various dynamic models used to estimate vehicle behavior, and the accuracy of each model can vary depending on driving conditions. Therefore, methods are being sought to more accurately estimate vehicle behavior in various driving situations. Figure 1 is a diagram illustrating the behavioral state of a vehicle. FIG. 2 is a drawing showing a vehicle behavior determination device according to an embodiment of the present invention. FIG. 3 is a drawing for explaining a method for determining the behavior of a vehicle according to an embodiment of the present invention. FIG. 4 is a diagram illustrating the input and output of a lateral velocity estimation unit and a roll/pitch estimation unit according to an embodiment of the present invention. Figure 5 is a schematic diagram showing the structure of an AI model according to an embodiment of the present invention. FIG. 6 is a diagram illustrating a method for learning a first weight according to an embodiment of the present invention. FIG. 7 is a diagram illustrating a method for learning second weights and third weights according to an embodiment of the present invention. FIGS. 8 to 10 are drawings for explaining a weight learning method according to an embodiment of the present invention. FIG. 11 is a drawing showing a computing system according to one embodiment of the present invention. Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the embodiments of the present invention, if it is determined that a detailed description of related known components or functions would hinder understanding of the embodiments of the present invention, such detailed description is omitted. In describing the components of the embodiments of the present invention, terms such as first, second, A, B, (a), (b), etc., may be used. These terms are intended merely to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the terms. Furthermore, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. 1 to 11. FIG. 1 is a drawing for explaining the behavioral state of a vehicle, and FIG. 2 is a drawing showing a vehicle behavior determination device according to an embodiment of the present invention. Referring to FIGS. 1 and 2, a vehicle behavior determination device according to an embodiment of the present invention may include a sensor (10), a processor (100), and a memory (20). The sensor (10) may be for acquiring one or more dynamic characteristics of a vehicle and may include a plurality of sensors for acquiring different physical quantities. The dynamic characteristics of a vehicle referred to in this specification may mean sensor data acquired by the sensor (10). For example, the dynamic characteristics of a vehicle may include steering angle, longitudinal angle, lateral acceleration, and yaw rate. The sensor (10) may include a steering angle sensor that measures the steering angle of the vehicle, a wheel speed sensor that measures the wheel speed, et