A method for calculating torque when a bidirectional shuttle bus changes lanes

Abstract

The present invention belongs to the field of automobiles, and specifically relates to a method for calculating torque when a two-way shuttle bus changes lanes. The method can solve the desired acceleration based on the reference speed, longitudinal speed, and acceleration based on model predictive control, and after obtaining the desired acceleration, the required driving torque or braking torque is calculated through the vehicle model. The present invention can solve the problem that the current automatic driving does not accurately calculate the vehicle driving or braking torque during the lane change process, and has the characteristics of accurate calculation results, conforming to the actual situation, and being conducive to vehicle control during lane change.

Description

[0001] This invention application is a divisional application based on the Chinese patent No. 2021102942727 submitted by the applicant on March 19, 2021, and the patent name is "A control system for a two-way shuttle bus and a two-way shuttle bus". Technical Field

[0002] The invention belongs to the field of automobiles, and in particular relates to a method for calculating torque when a bidirectional shuttle bus changes lanes. Background Art

[0003] A driverless car is a type of smart car, also known as a wheeled mobile robot, which mainly relies on an intelligent driver's instrument based on a computer system inside the car to achieve the purpose of autonomous driving.

[0004] Existing autonomous driving technology mainly relies on lidar and monocular cameras for image recognition and obstacle detection to achieve autonomous driving, but the current autonomous driving does not accurately calculate the vehicle's driving or braking torque during lane changes. Summary of the invention

[0005] In order to solve the above technical problems, an embodiment of the present invention proposes a method for calculating torque when a bidirectional shuttle bus changes lanes.

[0006] A method for calculating torque when a bidirectional shuttle vehicle changes lanes, wherein when the vehicle changes lanes, the driving or braking torque of the vehicle is calculated and determined, including:

[0007] Get the reference speed v ref , longitudinal velocity v, acceleration a;

[0008] The expected acceleration a of the vehicle is expressed as a first-order delay function des The relationship between it and the actual acceleration a is:

[0009]

[0010] Where K = 1 is the system gain, τ is the time constant, and s is the Laplace operator. The state equation of the continuous system can be expressed as:

[0011]

[0012]

[0013] The longitudinal control state equation is:

[0014] v(k+1)=v(k)+T·a(k)

[0015]

[0016] Where k is the current sampling time, k+1 is the next sampling time, T is the sampling period, let

[0017] ξ(k)=[v(k) a(k)] T

[0018] The control state equation can be expressed as:

[0019]

[0020] The goal of longitudinal control is to accurately track the reference speed without causing excessive acceleration and speed changes in the vehicle. The objective function can be defined as:

[0021]

[0022] Where P and Q are the weight matrices of the system control increment and the system output, respectively; Np is the prediction step length; and Nc is the control step length;

[0023] The system constraints are:

[0024] a des，min ≤a des (k+i)≤a des，max

[0025] Δa des，min ≤Δa des (k+i)≤Δa des，max

[0026] The optimal expected acceleration a can be obtained by minimizing the function J. des ;

[0027] The resistance of vehicle movement is the tire rolling resistance F roll , air resistance F a , slope resistance F grade , the drag acceleration is:

[0028]

[0029] When the required acceleration a des Greater than -a thre , driving force needs to be applied, when the required acceleration a des Less than -a thre It is necessary to apply braking force. To avoid frequent mode switching, a certain hysteresis is applied. The torque can be expressed as:

[0030]

[0031] The technical solution adopted by the present invention has the following beneficial effects:

[0032] Can be based on the reference speed v ref , longitudinal velocity v, acceleration a, and achieve the desired acceleration a based on model predictive control des Solving for the expected acceleration a des The required driving torque or braking torque is then calculated through the vehicle model. The calculation result is accurate and consistent with the actual situation, which is beneficial to vehicle control when changing lanes.

[0033] The specific technical solutions and beneficial effects of the present invention will be described in detail in the following specific embodiments in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments:

[0035] Figure 1 This is a schematic structural diagram of a control system for a bidirectional shuttle bus according to a first embodiment of the present invention;

[0036] Figure 2 This is a flow chart of the first processing module of the first embodiment of the present invention;

[0037] Figure 3 Schematic diagram of vehicle travel coordinates in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Embodiment 1

[0040] The basic idea of ​​this embodiment is to respectively set a camera module and an obstacle detection module at the front and rear ends of the vehicle, and determine first vehicle driving information through a first processing module according to the vehicle front image sent by the first camera module and the vehicle front end obstacle data sent by the first obstacle detection module, and send the first vehicle driving information to a power controller; and determine second vehicle driving information through a second processing module according to the vehicle front image sent by the second camera module and the vehicle rear end obstacle data sent by the second obstacle detection module, and send the second vehicle driving information to the power controller, thereby significantly improving the steering and lane changing flexibility of the vehicle.

[0041] It should be noted that the control system can be applied to a four-wheel drive shuttle bus, and can also be applied to other bidirectional multi-wheel drive shuttle buses, such as an eight-wheel drive shuttle bus.

[0042] Figure 1 This is a schematic diagram of the structure of a control system for a bidirectional shuttle bus provided in Embodiment 1 of the present invention. Figure 1 As shown, a control system for a bidirectional shuttle bus includes a first camera module for acquiring images in front of the vehicle, a second camera module for acquiring images behind the vehicle, a first processing module, a second processing module, a first obstacle detection module for detecting obstacle data at the front end of the vehicle, a second obstacle detection module for detecting obstacle data at the rear end of the vehicle, a power controller and a motor driver.

[0043] The first processing module is used to determine the first vehicle driving information based on the vehicle front image sent by the first camera module and the vehicle front obstacle data sent by the first obstacle detection module, and send the first vehicle driving information to the power controller. The second processing module is used to determine the second vehicle driving information based on the vehicle front image sent by the second camera module and the vehicle rear obstacle data sent by the second obstacle detection module, and send the second vehicle driving information to the power controller. The power controller is used to generate the first vehicle driving instruction and the second vehicle driving instruction according to the first vehicle driving information and the second vehicle driving information, and send them to the motor driver. The motor driver is used to drive the corresponding front wheels or rear wheels according to the first vehicle driving instruction and the second vehicle driving instruction.

[0044] The first processing module determines the first vehicle driving information based on the front image of the vehicle sent by the first camera module and the front obstacle data of the vehicle sent by the first obstacle detection module, and sends the first vehicle driving information to the power controller. The second processing module determines the second vehicle driving information based on the front image of the vehicle sent by the second camera module and the rear obstacle data of the vehicle sent by the second obstacle detection module, and sends the second vehicle driving information to the power controller. This control system significantly improves the steering and lane changing flexibility of the vehicle by controlling the movement of the front and rear wheels.

[0045] In this embodiment, the first processing module is connected to the first camera module, the first obstacle detection module and the power controller, the second processing module is connected to the second camera module, the second obstacle detection module and the power controller, the power controller is connected to the motor driver, and the motor driver is connected to the corresponding wheel.

[0046] In this embodiment, the first obstacle detection module includes a millimeter wave radar module and an ultrasonic radar module. The second obstacle detection module includes a millimeter wave radar module and an ultrasonic radar module.

[0047] Millimeter waves are essentially electromagnetic waves. The frequency band of millimeter waves is quite special. Its frequency is higher than radio and lower than visible light and infrared. The frequency range is approximately 10GHz-200GHz, which is a very suitable frequency band for the automotive field. There are six millimeter wave radar modules, three of which are located at the front and rear ends of the vehicle respectively. It should be noted that the first obstacle detection module can also use other radar modules, and this embodiment does not limit them.

[0048] The ultrasonic transmitter emits an ultrasonic signal in a certain direction outside, and starts timing at the moment of ultrasonic emission. The ultrasonic wave propagates through the air, and will immediately reflect and propagate back if it encounters an obstacle during propagation. The ultrasonic receiver stops timing immediately when it receives the reflected wave. The propagation speed of ultrasonic waves in the air is 340m / s. By recording the time t, the timer can measure the distance (s) from the emission point to the obstacle. There are eight ultrasonic radar modules, and the front and rear four are respectively located at the front and rear ends of the vehicle. It should be noted that the second obstacle detection module can also use other radar modules, and this embodiment does not limit them.

[0049] In order to achieve precise control of the vehicle, the first processing module is specifically used to: perform 3D-FFT processing on the obstacle data at the front of the vehicle from the first obstacle detection module to obtain a polar coordinate image, and then obtain two sizes of first feature images in the rectangular coordinate system through polar coordinate to Cartesian coordinate transformation and convolution feature extraction; obtain an image corresponding to the rectangular coordinate system of the first obstacle detection module through inverse projection mapping according to the joint calibration parameters with the first obstacle detection module, and then obtain a second feature image of the same size as the first feature image through convolution feature extraction; the first feature image and the second feature image of the same size will be used as the superimposed feature map of the two channels for feature extraction through the convolution kernel, and the output will be used as the input of the single-shot detector for target perception; predict the target motion trajectory through Kalman filtering, combine the obstacle data at the front of the vehicle with the global path planning to determine the current local path, and finally determine the current vehicle longitudinal speed and slew rate; calculate and determine the vehicle driving or braking torque, steering wheel angle and instantaneous angular velocity based on the current vehicle longitudinal speed and slew rate combined with the vehicle model. The information from the ultrasonic radar in the first obstacle detection module is used by the first processing module to detect close-range obstacles as a redundant safety measure to avoid collisions.

[0050] like Figure 2 As shown in an example embodiment, the radar raw data is processed by 3D-FFT, coordinate transformation and convolution feature extraction to obtain feature map 0 (32*32) and feature Figure 1 (64*64), the camera image data joint calibration parameters are inversely mapped and projected and convolution feature extracted to obtain feature map 0 (32*32) and feature Figure 1 (64*64), the feature map 0 (32*32) corresponding to the radar raw data and the feature map 0 (32*32) corresponding to the camera image data are superimposed and then 1*1 convolution is performed. The feature map corresponding to the radar raw data Figure 1 (64*64) and the features corresponding to the camera image data Figure 1 After feature superposition, (64*64) is subjected to 1*1 convolution processing, and the processing results of the two are used as the input of the SSD single-shot detector for target perception.

[0051] In order to achieve precise control of the vehicle, the second processing module is specifically used to: perform 3D-FFT processing on the obstacle data at the rear end of the vehicle from the second obstacle detection module to obtain a polar coordinate image, and then obtain two sizes of first feature images in the rectangular coordinate system through polar coordinate to Cartesian coordinate transformation and convolution feature extraction; obtain an image corresponding to the rectangular coordinate system of the second obstacle detection module through inverse projection mapping according to the joint calibration parameters with the second obstacle detection module, and then obtain a second feature image of the same size as the first feature image through convolution feature extraction; the first feature image and the second feature image of the same size will be used as the superimposed feature map of the two channels for feature extraction through the convolution kernel, and the output will be used as the input of the single-shot detector for target perception; predict the target motion trajectory through Kalman filtering, combine the obstacle data at the front end of the vehicle with the global path planning to determine the current local path, and finally determine the current vehicle longitudinal speed and slew rate; calculate and determine the vehicle driving or braking torque, steering wheel angle and instantaneous angular velocity based on the current vehicle longitudinal speed and slew rate combined with the vehicle model. The information from the ultrasonic radar in the second obstacle detection module is used by the second processing module to detect close-range obstacles as a redundant safety measure to avoid collisions.

[0052] The data processing method of the second processing module is the same as that of the first processing module, so it will not be repeated here.

[0053] In an embodiment, the reference speed v is obtained ref , longitudinal velocity v, acceleration a, and achieve the desired acceleration a based on model predictive control des Solving for the expected acceleration a des The required driving or braking torque is then calculated using the vehicle model.

[0054] The expected acceleration a of the vehicle is expressed as a first-order delay function des The relationship between it and the actual acceleration a is:

[0055]

[0056] Where K = 1 is the system gain, τ is the time constant, and s is the Laplace operator. The state equation of the continuous system can be expressed as:

[0057]

[0058]

[0059] The longitudinal control state equation is:

[0060] v(k+1)=v(k)+T·a(k)

[0061]

[0062] Where k is the current sampling time, k+1 is the next sampling time, T is the sampling period, let

[0063] ξ(k)=[v(k) a(k)] T

[0064] The control state equation can be expressed as:

[0065]

[0066] The goal of longitudinal control is to accurately track the reference speed without causing excessive acceleration and speed changes in the vehicle. The objective function can be defined as:

[0067]

[0068] Where P and Q are the weight matrix of the system control increment and the weight matrix of the system output respectively, Np is the prediction step size, and Nc is the control step size.

[0069] The system constraints are:

[0070] a des，min ≤a des (k+i)≤a des，max

[0071] Δa des，min ≤Δa des (k+i)≤Δa des，max

[0072] The optimal expected acceleration a can be obtained by minimizing the function J. des .

[0073] The resistance of vehicle movement is the tire rolling resistance F roll , air resistance F a , slope resistance F grade , the drag acceleration is:

[0074]

[0075] When the required acceleration ades Greater than -a thre , driving force needs to be applied, when the required acceleration a des Less than -a thre It is necessary to apply braking force. To avoid frequent mode switching, a certain hysteresis is applied. The torque can be expressed as:

[0076]

[0077] According to the reference trajectory and the current vehicle lateral state variables, the front and rear wheel steering angles are solved based on the vehicle kinematic model.

[0078] The vehicle is equivalent to a two-wheeled vehicle model, such as Figure 3 As shown, a rectangular coordinate system including the vehicle is established, where A is the center of the front axle, B is the center of the rear axle, C is the center of mass of the vehicle, β is the side slip angle of the vehicle, is the vehicle heading angle, δ f is the front wheel steering angle, δ r is the rear wheel steering angle, V is the velocity of the center of mass, R is the instantaneous turning radius, O is the center of the turning circle, l f is the distance from the center of mass to the front axle, l r are the distances from the center of mass to the rear axle.

[0079] In triangles OCA and OCB we have:

[0080]

[0081]

[0082] The sideslip angle can be obtained from the above two equations:

[0083]

[0084] The instantaneous turning radius is:

[0085]

[0086] Yaw rate:

[0087]

[0088] The X and Y velocities of the center of mass are:

[0089]

[0090]

[0091] Define the discrete model control quantity as u L =|δ f δ r ] T, the output is The reference position and actual position of the center of mass are Under the ideal reference model, the vehicle sideslip angle β=0.

[0092] The instantaneous turning center of the vehicle is:

[0093]

[0094]

[0095]

[0096] Based on the center of the circle, the discrete state of the system is:

[0097]

[0098]

[0099]

[0100] The objective function is defined as:

[0101]

[0102] where y L，p (k+i|k) is the predicted value of the output variable, y L，ref (k+i|k) is the reference value of the output variable, S is the weight matrix of the system output, and W is the weight matrix of the system control quantity.

[0103] The system constraints are the limit values ​​of the front and rear wheel steering angles and their increments.

[0104] u L，min ≤u L (k+i)≤u L，max

[0105] Δu L，min ≤Δu L (k+i)≤Δu L，max

[0106] In addition, the objective cost function takes the minimum value to obtain u L =|δ f δ r ] T .

[0107] Assuming the ratio of the steering wheel angle θ to the tire angle δ is n, we have:

[0108] θ(k)=n*δ(k)

[0109] The instantaneous steering wheel speed is:

[0110]

[0111] Where T is the sampling time.

[0112] by u L The steering wheel angle and instantaneous angular velocity can be calculated.

[0113] During the vehicle's forward or backward movement, the control system works as follows: when the vehicle needs to move forward, the first camera module sends a positive torque request to the power controller, and after the power controller recognizes the first camera module, it assigns the corresponding first vehicle driving information to the corresponding motor driver, thereby making the vehicle move forward; when the vehicle needs to move backward, the second camera module sends a positive torque request to the power controller, and after the power controller recognizes the second camera module, it assigns the corresponding second vehicle driving information to the corresponding motor driver, thereby making the vehicle move backward.

[0114] Preferably, the first camera module is connected to the power controller via a CAN network, and after the power controller identifies the CANID of the first camera module, it distributes the corresponding torque value to the corresponding multiple motor drivers to achieve precise control of the motor drivers. The same is true for the second camera module, so it will not be described in detail.

[0115] During the vehicle braking process, the control system works as follows: when the vehicle is moving forward and performing energy recovery braking, the first camera module sends a negative torque request, which is sent to the motor driver through the power controller to perform energy recovery braking; when the vehicle is moving backward and performing energy recovery braking, the second camera module sends a negative torque request, which is sent to the motor driver through the power controller to perform energy recovery braking.

[0116] When the vehicle changes lanes, the power controller controls the motor driver to drive the front wheels and swing the rear wheels in the same direction to assist in changing lanes, thereby achieving rapid lane changes.

[0117] When the vehicle turns, the power controller controls the motor driver to drive the front wheels and swing the rear wheels in the opposite direction to reduce the turning radius.

[0118] Embodiment 2

[0119] This embodiment provides a bidirectional shuttle bus, including the control system of the bidirectional shuttle bus described in the first embodiment.

[0120] It can be seen from the description of Example 1 that this Example 2 proposes a two-way shuttle bus, in which a camera module and an obstacle detection module are respectively arranged at the front and rear ends of the vehicle, and a first processing module is used to determine the first vehicle driving information according to the vehicle front image sent by the first camera module and the vehicle front end obstacle data sent by the first obstacle detection module, and the first vehicle driving information is sent to the power controller; a second processing module is used to determine the second vehicle driving information according to the vehicle front image sent by the second camera module and the vehicle rear end obstacle data sent by the second obstacle detection module, and the second vehicle driving information is sent to the power controller, thereby significantly improving the steering and lane changing flexibility of the vehicle.

[0121] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes but is not limited to the contents described in the above specific embodiments. Any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. A method for calculating torque when a bidirectional shuttle bus changes lanes, characterized in that: When the vehicle changes lanes, the vehicle driving or braking torque is calculated and determined, including: Get the reference speed v ref , longitudinal velocity v, acceleration a; The expected acceleration a of the vehicle is expressed as a first-order delay function des The relationship between it and the actual acceleration a is: Where K = 1 is the system gain, τ is the time constant, and s is the Laplace operator. The state equation of the continuous system can be expressed as: The longitudinal control state equation is: v(k+1)=v(k)+T·a(k) Where k is the current sampling time, k+1 is the next sampling time, T is the sampling period, let ξ(k)=[v(k) a(k)] T The control state equation can be expressed as: The goal of longitudinal control is to accurately track the reference speed without causing excessive acceleration and speed changes in the vehicle. The objective function can be defined as: Where P and Q are the weight matrices of the system control increment and the system output, respectively; Np is the prediction step length; and Nc is the control step length; The system constraints are: a des，min ≤a des (k+i)≤a des，max Δa des，min ≤Δa des (k+i)≤Δa des，max The optimal expected acceleration a can be obtained by minimizing the function J. des ; The resistance of vehicle movement is the tire rolling resistance F roll , air resistance F a , slope resistance F grade , the drag acceleration is: When the required acceleration a des Greater than -a thre , driving force needs to be applied, when the required acceleration a des Less than -a thre Braking force needs to be applied. To avoid frequent mode switching, a certain hysteresis is applied. The torque can be expressed as:

Images