WO2023091104A1

WO2023091104A1 - A method for energy efficient control of active and semi-active suspension systems

Info

Publication number: WO2023091104A1
Application number: PCT/TR2021/051366
Authority: WO
Inventors: Mert BÜYÜKKÖPRÜ; Xavier Mouton; Erdem UZUNSOY
Original assignee: Oyak Renault Otomobi̇l Fabri̇kalari Anoni̇m Şi̇rketi̇; Bursa Tekni̇k Üni̇versi̇tesi̇ Rektörlüğü
Priority date: 2021-11-19
Filing date: 2021-12-07
Publication date: 2023-05-25

Abstract

The invention is related to the method which that enables the active or semi-active suspension system to be controlled by measuring the road profile with the camera (20) by creating a virtual vehicle (30) with the same parameters to simulate the current vehicle controlled by at least one processor in the digital environment, ensuring the movement of the virtual vehicle (30) on the road profile which is calculated with the data received from the vehicle (10), calculation of suspension stroke speed, body speed and wheel deflection speed, which are the dynamic parameters of the virtual vehicle (30) during the said movement, calculation of forces and movements by creating a wv matrix for suspension control, updating gain coefficients from the virtual vehicle (30), optimizing the distribution of force movements, creating control outputs to control the vehicle's (10) suspension system.

Description

A METHOD FOR ENERGY EFFICIENT CONTROL OF ACTIVE AND SEMI-ACTIVE SUSPENSION SYSTEMS

FIELD OF THE INVENTION

The invention relates to a method that enables the management of active or semiactive suspension systems used to increase driving comfort and driving safety in vehicles.

In particular, the invention relates to a method that enables the determination of suspension control forces by making estimation in order to analyze the road characteristics beforehand while driving, and thus enables the efficient use of energy while performing the planned suspension control.

PRIOR ART

Road profile structures are one of the most important parameters that affect driving comfort and safety during driving a vehicle. The road structure may vary depending on the environmental conditions, both in urban and suburban roads. Particularly, the road condition on stabilized roads or on modified roads affects the driver significantly in terms of driving quality.

Cameras have been widely used with the developing technology in the automotive sector. Thus, fully autonomous or semi-autonomous driving has been developed and the vehicle can be safely driven independent of the driver. Analysis is realized by processing the environmental data received from the camera, and the driver can be warned with the road condition and/or the vehicle can be driven by itself.

In prior art applications, the road profile structure at a certain distance is detected by means of at least one camera on the vehicle and the instant state of the road is analyzed, and information is transferred from the control unit. In known applications, it is also possible to automatically control the suspension system, particularly by detecting the (recess-protrusion/pothole-bump) structures of altitude and lowness on the road. However, while said structure is scanned with the detector (camera) with the movement of the vehicle, there is noise on the profile obtained based on the parameters such as variable vehicle speed. Particularly, factors such as vibration on the vehicle also increase the noise. In this case, it causes the improper operation of the road profile calculation algorithms and the desired comfort/information cannot be provided to the user.

In prior art applications, especially an active suspension system, it is carried out instantly depending on the scanning of the road which the systems that provide automatic adjustment of the suspension level. In semi-active suspension systems, force produces depending on the disturbing input from the road. Damping rate can be adjusted to control the amount of force. The continual changes of the road in fully active suspensions causes high energy consumption of the suspension element. However, energy consumption in semi-active suspension elements is low. In known applications, there is no technic that provides preliminary analysis with the estimation of the road profile, especially in semi-active suspension systems.

Especially due to the problems described above, in case of noise occurring during road profile estimation, problems such as incorrect adjustment of the suspension or continuous energy usage encountered. In the prior art, there is no application that provides the virtual modeling of the road profile in order to adjust the control forces of the suspension system and the adjustment of the suspension control force depending on the results of this model.

An application that makes road profile estimation disclosed in the LISPTO patent document with the publication number US20150254803A1 . As explained above, the system can determine the direction of the road by performing the road profile on two axes by means of the camera. Parameters such as the depth on the road are realized by using a secondary sensor. As mentioned above, depth and height parameters cannot be analyzed due to noise.

As a result, all abovementioned problems have made it necessary to make an improvement in the relevant technical field. AIM OF THE INVENTION

The present invention aims to eliminate the abovementioned problems and to make a development in the relevant technical field.

The main objective of the invention is to determine the road profile characteristic by removing noise in anticipation while driving and to reveal the method that enables the active or semi-active suspension system to be effectively controlled by simulating the vehicle's suspension system on the estimated road profile.

Another objective of the invention it is to ensure that the energy used during the control of the suspension system is managed efficiently.

Another objective of the invention it is to ensure that reveal the energy-saving method on vehicles which have active or semi-active suspension system.

Another objective of the invention it is to ensure that save energy by eliminating the need for continuous control according to the stroke speed of semi-active suspension systems.

BRIEF DESCRIPTION OF THE INVENTION

The invention is related to a method for energy efficient control of active and semiactive suspension systems so as to fulfil all aims mentioned above and will be obtained from the following detailed description.

The camera captures pictures so as to detect the profile structure of the road within the scope of the system. The frequency of capturing pictures varies according to the application used. The points determined within the system are aligned in the regions other than the Z axis. The data set is obtained by using the obtained point cloud. The camera captures pictures so as to detect the profile structure of the road. The taken pictures are subjected to image processing. After the image processing process, road profile measurement points are generated. The determined time may vary in alternative applications. A data cloud is created with the obtained points.

During the successive measurement of the same point, a shift (sliding) occurs in the x- axis since the distance between two points in the point cloud is variable depending on the vehicle speed. As a result of the shifts, the distance between the points remains below the two-point measurement capacity of the camera. In this case, more effective results can be obtained by filtering and optimizing the generated noise and it is ensured to analyse the potholes and bumps on the road. The inventive method is used during the filtering process, because the high frequency path profile data is meaningful for the analysis detail. Thus, estimation can be made independent of the speed of the vehicle.

The invention is a method for detecting and processing surface shapes (in x-y-z axes) with road (R) profiles detected by a camera that acts as at least one sensor on vehicles, characterized by carrying out the following process steps; controlling whether the images taken by said camera contain height and distance (x-y axis) data or not, creating a point cloud for points, containing data points with height and distance, adding the data of the same surface shape to the point cloud, alignment of points in the point cloud, clustering of aligned points and determination of adjacent radius in the connection, filtering out the noise outside the cluster, converting data from space (distance, height, width) domain to time domain.

In a preferred embodiment of the invention, it is possible to determine the randomly distributed surface shapes on the driving route by means of the method.

In another preferred embodiment of the invention, while the vehicle detects the changes in the road profile detected by the camera while it is moving, density-based spatial clustering algorithm is used, which ensures that it is filtered without loss at high frequency.

In another preferred embodiment of the invention, conversion is made in the time domain according to vehicle speed, function (operation) time and road profile position.

In another preferred embodiment of the invention, noises are filtered out by filtering process caused by domain change after the time domain conversion and camera capturing measurement.

In another preferred embodiment of the invention, the noises resulting from the domain change and camera capturing measurement are filtered with a low-pass filter after the time domain conversion. In another preferred embodiment of the invention, the noises resulting from the domain change and camera capturing measurement are filtered with a high-pass filter after the time domain conversion.

In another preferred embodiment of the invention, the conversion of the camera position according to the Earth coordinate order is performed by using the distance between the vehicle's centre of gravity and the camera, and the distance between the vehicle's rear wheel-road contact point and the centre of gravity. In alternative applications, camera angle can be calculated by means of the built-in gyroscope of the camera.

In another preferred embodiment of the invention, the operation is performed without the need for permanent verification of the camera position by distinguishing the change in suspension height due to vehicle centre of gravity on inclined roads.

In another preferred embodiment of the invention, the process of transforming data from space plane (distance, height, width) domain to time domain is interpolation.

In another preferred embodiment of the invention, alignment is made so as to ensure that the same road profile data is verified without affecting each other.

The invention is related a method that enables the active or semi-active suspension system to be controlled by using the road profile data obtained by performing the operations in vehicles, to provide detection and processing of surface shapes (on x-y-z axes) with road profiles detection by a camera or lidar which acts as at least one sensor; and its features are;

• Checking the images taken from the said camera whether contain height and distance (x-y axis) data,

• Generating point cloud for points of containing data points with height and distance values,

• Adding the data to the point cloud which have same surface shape,

• Alignment of points in the point cloud,

• Clustering the aligned points and determining the adjacent radius in the connection, Filtering of noises outside the cluster,

Converting data from space plane (distance, height, width) domain to time domain characterized in that with bellow steps; a. Creating a virtual vehicle with the same parameters to simulate the current vehicle controlled by at least one processor in the digital environment, ensuring the movement of the virtual vehicle on the road profile which is calculated with the data received from the vehicle, b. Calculation of suspension stroke speed, body speed and wheel deflection speed, which are the dynamic parameters of the virtual vehicle during the said movement, c. Calculation of forces and movements by creating a wv matrix for suspension control, d. Updating gain coefficients from the virtual vehicle, e. Optimizing the distribution of force movements, f. Creating control outputs to control the vehicle’s suspension system.

In a preferred embodiment of the invention, the effectiveness of the camera is verified by measuring the stroke and acceleration data which are the data obtained from physical sensors with the vehicle passing over the calculated road profile and comparing them with the virtual vehicle’s sensors (virtual) produced for the same road profile.

In a preferred embodiment of the invention, if the measurements are verified or the accuracy is determined, the next optimization and the next road profiling are improving automatically.

In a preferred embodiment of the invention, the control outputs mentioned in the e process step which is optimizing the distribution of force movements; which can produce control outputs according to the driving mode, emergency detection and energy management parameters are produced over the data which are obtained by the camera and which are filtered for the road profile estimation. In a preferred embodiment of the invention, where the wv matrices are changed for the optimization of the virtual forces calculated with the virtual vehicle according to the driving mode, emergency detection, energy management parameters.

In a preferred embodiment of the invention, where the forces and movements are calculated by creating wv matrix in order to enable the suspension stroke rate to be determined independently for each wheel by activating the semi-active suspension system’s damper before the force effects on the damper.

In a preferred embodiment of the invention, the suspension members stroke speed on each wheel are measured to determine i. Pitch movement ii. Roll movement iii. Bounce movement

According to the virtual center of gravity on the virtual vehicle.

In a preferred embodiment of the invention, it optimizes the distribution of forces with the C.A. algorithm

The protection scope of the invention is specified in the claims and cannot be limited to the description made for illustrative purposes in this brief and detailed description. It is clear that a person skilled in the art can present similar embodiments in the light of the above descriptions without departing from the main theme of the invention.

BRIEF DESCRIPTION OF DRAWINGS

In Figure 1 , a representative drawing expressing the elements used in determining the camera coordinate within the scope of the invention is given.

In Figure 2, a representative drawing describing the operation of the invention is given.

In Figure 3, the flow chart which is describing the operation of the method that enables the suspension system to be controlled within the scope of the invention is given.

In Figure 4, the diagram describes the operation of the control system. In Figure 5 and Figure 6, the diagram describing how the invention makes a road profile estimation is given.

In Figure 7, a drawing describing the alignment in the point cloud is given.

In Figure 8, an example showing the operation of the system which is activated an emergency situation is given.

In Figure 9, a graph showing the force comparison during low and high level control is given.

In Figure 10, a graph showing the comparison of course speed and electric current is given.

DESCRIPTION OF THE REFERENCES IN FIGURES

10. Vehicle

11 . Center of gravity

20. Camera

30. Virtual vehicle

31 . Virtual vehicle center of gravity

R. Road

DETAILED DESCRIPTION OF THE INVENTION

In this detailed description, the inventive a method for energy efficient control of active and semi-active suspension systems is described by means of examples only for clarifying the subject matter such that no limiting effect is created.

In particular, the invention relates to a method that enables the determination of suspension control forces by making estimation in order to analyze the road characteristics beforehand while driving, and thus enables the efficient use of energy while performing the planned suspension control. The invention is a method that enables the active or semi-active suspension system to be controlled by using the road profile data obtained by performing the operations in vehicles (10), to provide detection and processing of surface shapes (on x-y-z axes) with road (R) profiles detection by a camera (20) or lidar which acts as at least one sensor; and its features are;

• Checking the images taken from the said camera (20) whether contain height and distance (x-y axis) data,

• Adding the data to the point cloud which have same surface shape,

• Alignment of points in the point cloud,

• Clustering the aligned points and determining the adjacent radius in the connection,

• Filtering of noises outside the cluster,

• Converting data from space plane (distance, height, width) domain to time domain characterized in that with bellow steps; g. Creating a virtual vehicle (30) with the same parameters to simulate the current vehicle controlled by at least one processor in the digital environment, ensuring the movement of the virtual vehicle (30) on the road profile which is calculated with the data received from the vehicle (10), h. Calculation of suspension stroke speed, body speed and wheel deflection speed, which are the dynamic parameters of the virtual vehicle (30) during the said movement, i. Calculation of forces and movements by creating a wv matrix for suspension control, j. Updating gain coefficients from the virtual vehicle (30), k. Optimizing the distribution of force movements, l. Creating control outputs to control the vehicle’s (10) suspension system. In the preferred embodiment of the invention, the suspension members stroke speed on each wheel are measured to determine forces relative to the virtual center of gravity (31 ) on the virtual vehicle (30) with these parameters; i. Pitch movement which is rotation on the transverse axis. ii. Roll movement which is rotation on the longitudinal axis. iii. Bounce movement which is rotation on the vertical axis.

Within the method of the invention, an image is captured from the road (R) by means of the camera (20), which acts as a sensor. The camera (20) located in the front part of the vehicle (10), preferably on the windshield. In this context, the method of the invention received data are processed with optimized.

In Figure 1 and Figure 2, drawings describes the running of the invention. As seen in Figure 1 , during the progress of the vehicle (10) on the straight road (R), the road (R) is scanned by the camera (20) and the clusters of points of the height parameters are collected detected by the camera (20). The representative diagram of the system given in Figure 6.

Within the scope of the invention, the collected data are on the X, Y and Z axis based from Earth coordinate system and the data measured as transformed according to the X, Y and Z axis of the camera (20). In this way, data cloud created which are containing distance and altitude data. In case, if each data coming from the camera (20) contains distance and height data, it is added into the data cloud. Said data is added depending on the speed of the vehicle (10) and the data flow frequency is affected by the speed of the vehicle (10). In this context, same data is obtained more than once while the vehicle (10) is moving, and these data separated by comparing them with the existing data. Firstly, the points are aligned and the noise is filtered.

In this context, the data points which are collected in different periods must be aligned with each other. It is important whether the data points received are empty (whether they contain X-Y-Z coordinate data). The created points are weighted, and the weight is the reliability value of the measured height data. The fracture regions of the measured data constitute the maximum and minimum parts of the height data. Thus, the data are clustered for each road (R) profile and the noise outside the cluster is filtered. Closer points in the horizontal direction that are less than the theoretical minimum distance between two points can result in point flakes. In terms of profile evaluation that takes into account relatively high frequencies, point flakes excess produces artificial high frequency content as a result of data processing.

Within the scope of the invention, data sets are interpolated and transferred in the time domain. In this way, the data received from the camera (20) can be examined on a time basis together with the communication time on the CAN line of the vehicle (10).

The predicted road (R) profile is captured with the camera (20) on the distance domain. During the mentioned process, the function time, speed and profile height are taken into consideration. Integration with vehicle speed is performed by means of linear interpolation.

In the process of transferring the said data from the distance domain to the time domain, it is an important criterion to determine the position of the camera (20) with respect to the Earth coordinate.

Within the scope of the invention, the noises which are generated on the road (R) profile data converted to time domain filtered with high-pass and low-pass filters.

Within the scope of the invention, each road (R) profile data can be separated and evaluated independently during the clustering process. The points which are locate outside the cluster defined as noise and the noises in the data are filtered. In this way, it is possible to examine the road (R) profile data containing distance and height data in higher resolution without losing data.

Within the scope of the invention, the distance of the camera (20) (sensor) from the center of gravity (12) and the distance of the center of gravity (12) to the rear wheelroad contact point (11) are used during the transformation of the camera (20) on the vehicle (10) on the time base according to the world coordinate system. In this way, road (R) profile locations can be determined over the camera (20) location. Considering the representative positions indicated in Figure 1 , the vehicle (10) suspension state of the camera (20) position or the position against the reverse forces taken by the road (R) can be taken into consideration and the instantaneous position of the camera (20) in this situation can be evaluated. Since the changes of the center of gravity (12) of the vehicle (10) according to the suspension situation will affect the result, the mentioned parameters are taken into consideration while making the conversion from the distance domain to the time domain. In the preferred embodiment of the invention, the algorithm used during the clustering process is the density-based spatial clustering algorithm.

In the preferred embodiment of the invention, after the conversation process in the time domain, the noise which is arising from the conversion and camera (20) measurement which couldn’t be filtered in the previous processes is filtered with a high-pass filter and/or low-pass filter so that the frequency is not affected.

In Figure 4, the drawing describes the operation of the active and semi-active suspension system control method, which is the subject of the invention. After the road profile is calculated by the camera (20), the road (R) profile is created virtually and the virtual vehicle (30) is moved on the road (R) profile. Depending on the mentioned movement, preliminary analysis is made and according to the results of this analysis, it is possible to control the active or semi-active suspension system. The dynamic parameters of the virtual vehicle (30) are primarily calculated on the profile. Said dynamic parameters are the stroke speed of each suspension component, the body speed of the virtual vehicle (30) and the wheel speed of the virtual vehicle (30). The system can be classified as high-level or low-level suspension control. Mentioned the motion parameters are determined together with the planning of the control gains. As a result of the analysis, the distribution of force movements is optimized. In this context, control output is produced based on damper optimization, driving mode, emergency and energy management. By controlling the driving mode, the wv matrices are preferably changed and the force distribution is optimized for the optimization of the virtual forces. C.A algorithm is preferably used for optimization. The control output can be for the semi-active suspension system’s damper or the active suspension system, depending on the equipment of the vehicle (10).

In particular, when the semi-active suspension system is considered, it is aimed to save energy by determining the suspension stiffness based on the dynamic analysis made beforehand, rather than instantaneously.

As a result of the vehicle (10) passing over the mentioned road (R) profile, the outputs captured from the physical sensors on the vehicle (10) and the calculated outputs of the virtual vehicle (30) are compared. If the measurements are confirmed, the next virtual optimization is improved. Similarly, if the physical outputs are compared with the camera (20) data and the effectiveness of the calculated outputs is verified, the next road (R) profile is developed by using the obtained parameters.

In the preferred embodiment of the invention, while the profile is created with the virtual vehicle (30), the height difference between the front and rear wheels of the vehicle (10) is analyzed depending on the profile. In this way, it is possible to make verification and preliminary analysis while creating the road (R) profile.

In the preferred embodiment of the invention, the road (R) profile response is analyzed with the virtual vehicle (30), while the virtual body speed in the Z axis, the virtual suspension travel speed and the rotational movements of the vehicle in the Z axis are analyzed. These parameters which are the angles of hitting, rolling, turning from the center (pitch, roll, bounce/yaw) are based on the virtual vehicle center of gravity (31 ).

Emergency Conditions

In case of emergency situations, such as cornering or overturning of the vehicle, these forces must be minimized. An example describing the road condition is given in Figure 8. The road (R) profile seen in the lower part of the figure comes on the right wheels, while the road (R) profile in the upper part comes on the left wheels. In the example given in the figure, the suspension system on the left is activated more effectively (semi-active/active suspension), allowing the vehicle (10) to travel safely on the road (R).

The force-velocity characteristic curve in semi-active suspension systems is non-linear. The force is controlled by electric current c(i) and varies according to the stroke rate z_s - z_u. The equations expressing the current characteristic are given below.

The variable definitions used in the calculation equations are given in Table 1 .

Table 1 The matrix to be used in the control calculation;

And force control equations;

And matrix output for calculation of forces;

As above.

In the semi-active suspension system;

The equation for the calculation of the data taken from the road (R) profiles is given below. W_v and/or W_u matrices used in the calculation;

As defined.

Transverse gain (Pitch); gain in rotational motion of the vehicle (10) on the transverse axis, longitudinal gain (Roll); gain in rotational motion of the vehicle (10) on the longitudinal axis, vertical gain (Bounce); is the gain in the rotational motion of the vehicle on the vertical axis.

Claims

1. A method that enables the active or semi-active suspension system to be controlled by using the road profile data obtained by performing the operations in vehicles (10), to provide detection and processing of surface shapes (on x-y-z axes) with road (R) profiles detection by a camera (20) or lidar which acts as at least one sensor; and its features are;

• Adding the data to the point cloud which have same surface shape,

• Alignment of points in the point cloud,

• Filtering of noises outside the cluster,

• Converting data from space plane (distance, height, width) domain to time domain characterized in that with bellow steps; m. Creating a virtual vehicle (30) with the same parameters to simulate the current vehicle controlled by at least one processor in the digital environment, ensuring the movement of the virtual vehicle (30) on the road profile which is calculated with the data received from the vehicle (10), n. Calculation of suspension stroke speed, body speed and wheel deflection speed, which are the dynamic parameters of the virtual vehicle (30) during the said movement, o. Calculation of forces and movements by creating a wv matrix for suspension control, p. Updating gain coefficients from the virtual vehicle (30), q. Optimizing the distribution of force movements, r. Creating control outputs to control the vehicle’s (10) suspension system.

2. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 1 , characterized in that;

The effectiveness of the camera (20) is verified by measuring the stroke and acceleration data which are the data obtained from physical sensors, with the vehicle (10) passing over the calculated road profile and comparing them with the virtual vehicle’s (30) sensors (virtual) produced for the same road profile.

3. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 2, characterized in that; if the measurements are verified or the accuracy is determined, the next optimization and the next road profiling are improves automatically.

4. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 1 , characterized in that; the control outputs mentioned in the e process step; which can produce control outputs according to the driving mode, emergency detection and energy management parameters are produced over the data which are obtained by the camera (20) and which are filtered for the road profile estimation.

5. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 4, characterized in that; where the wv matrices are changed for the optimization of the virtual forces calculated with the virtual vehicle (30) according to the driving mode, emergency detection, energy management parameters.

6. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 1 , characterized in that; where the forces and movements are calculated by creating wv matrix in order to enable the suspension stroke rate to be determined independently for each wheel by activating the semi-active suspension system’s damper before the force effects on the damper. Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 1 , characterized in that; the suspension members stroke speed on each wheel are measured to determine i. Pitch movement ii. Roll movement iii. Bounce movement

According to the virtual center of gravity (31 ) on the virtual vehicle (30). Active or semi-active suspension system to be control method based from estimated road (R) profile according to claim 1 , characterized in that; which optimizes the distribution of ferees with the C.A. algorithm.