A method, system and apparatus for road preview considering steering operation of wheeled vehicles

Abstract

The application discloses a road surface preview method, system and equipment considering steering operation of a wheeled vehicle, and relates to the technical field of automobile suspension control. The method comprises the following steps: when a steering wheel deflection angle is detected, acquiring point cloud data of a road surface in front of the wheeled vehicle; based on the point cloud data of the road surface in front of the wheeled vehicle, a three-dimensional map of the road surface in front of the vehicle wheel is constructed; according to the steering wheel deflection angle and the vehicle speed, a pre-travel path of each vehicle wheel in the wheeled vehicle is determined; according to a preview time and the pre-travel path of each vehicle wheel, a preview point of each vehicle wheel is determined; and elevation data of a preview area corresponding to the preview point of each vehicle wheel is extracted from the three-dimensional map. The application determines the preview point of each vehicle wheel by constructing the three-dimensional map of the road surface in front of the vehicle wheel, can realize road surface preview during steering of the vehicle, and further improves the control precision of the wheeled vehicle.

Description

Technical Field

[0001] This invention relates to the field of automotive suspension control technology, and in particular to a road surface pre-aiming method, system, and device that takes into account the steering operation of wheeled vehicles. Background Technology

[0002] Road surface preview technology improves the performance of a controllable suspension system by detecting road surface irregularities that the vehicle's wheels are about to traverse and using this information as feedforward input. Existing road surface preview methods include front-wheel preview and inter-axle preview. Front-wheel preview uses sensors at the front of the vehicle to collect road surface irregularities, which serve as control reference signals for both the front and rear suspension control systems. Front-wheel preview allows the suspension control system to anticipate upcoming road surface stimuli, resulting in better control. However, this method requires dedicated road surface preview sensors, leading to higher costs. Inter-axle preview uses the front wheels to sense road surface irregularities, using this information as a control reference signal for the rear suspension control system. This method only requires adding displacement sensors to measure the compression of the front suspension, thus reducing costs. However, inter-axle preview can only be used for rear suspension control, and the accuracy of the road surface irregularities obtained from the preview is not high. Furthermore, inter-axle preview is only suitable for straight-line driving conditions and is not applicable to turning driving conditions.

[0003] Sensors used for road surface prediction mainly fall into two categories: radar and cameras. Radar primarily includes mechanical lidar and solid-state lidar, while cameras are mainly categorized as monocular cameras, binocular cameras, and depth cameras. Because cameras are susceptible to environmental interference, solid-state lidar is often chosen for road surface prediction sensors in complex outdoor environments. The basic method involves using point cloud information of the road surface in front of the vehicle collected by the road surface prediction sensor and the vehicle's real-time status information to perform three-dimensional real-time localization and mapping of the road surface in front of the wheels. The elevation information of the local road surface that the wheels will contact at a future moment is then calculated based on the vehicle speed and used as input excitation for the suspension control system. The width of the local road surface extracted for prediction is generally slightly larger than the width of the wheel. The distance from the extracted local road surface to the wheel should be determined based on the vehicle speed and the response time of the electronic control system. For straight-line driving, the extracted local road surface only needs to be directly in front of the wheels. However, for turning driving, due to the sideslip angle, the wheel's rolling trajectory will not pass directly in front of the extracted local road surface. Therefore, it is necessary to propose a road surface prediction method that takes into account the steering operation of wheeled vehicles, so as to correctly predict the road surface that the wheels are about to pass under steering conditions. Summary of the Invention

[0004] The purpose of this invention is to provide a road surface prediction method, system, and device that takes into account the steering operation of wheeled vehicles, which can improve the road surface prediction accuracy and thus improve the control accuracy of wheeled vehicles.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] A road surface pre-aiming method considering steering operation of a wheeled vehicle, the method being applied to a wheeled vehicle equipped with multiple solid-state lidars;

[0007] The solid-state lidar is used to acquire point cloud data of the road surface in front of the wheeled vehicle;

[0008] The method includes:

[0009] When the steering wheel deflection angle is detected, point cloud data of the road surface in front of the wheeled vehicle is acquired;

[0010] A 3D map of the road surface in front of the wheels is constructed based on point cloud data of the road surface in front of the wheeled vehicle.

[0011] Obtain the speed of wheeled vehicles;

[0012] Based on the steering wheel deflection angle and the vehicle speed, determine the pre-driving path of each wheel in the wheeled vehicle;

[0013] Based on the aiming time and the pre-driving path of each wheel, determine the aiming point of each wheel;

[0014] Elevation data of the aiming area corresponding to the aiming point of each wheel are extracted from the 3D map.

[0015] Optionally, after extracting the elevation data of the aiming area corresponding to the aiming point of each wheel from the 3D map, the method further includes:

[0016] The elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel is used as the real-time input of the road excitation and input into the suspension control system of the wheeled vehicle.

[0017] Optionally, constructing a 3D map of the road surface in front of the wheels based on point cloud data of the road surface in front of the wheeled vehicle includes:

[0018] The point cloud data of the road surface in front of the wheeled vehicle were filtered to obtain multiple sets of filtered point cloud data.

[0019] The point cloud data after multiple filtering processes are downsampled to obtain multiple sets of downsampled point cloud data.

[0020] Feature points were extracted from each group of downsampled point cloud data to obtain the feature points of each group of downsampled point cloud data.

[0021] Based on the feature points, coordinate registration is performed on multiple sets of downsampled point cloud data, and the multiple sets of downsampled point cloud data are transformed into the same coordinate system to obtain a three-dimensional map of the road surface in front of the wheel.

[0022] Optionally, based on the steering wheel deflection angle and the vehicle speed, the pre-travel path of each wheel in the wheeled vehicle is determined, including:

[0023] Obtain the current steering ratio of the inner wheel and the steering ratio of the outer wheel.

[0024] Based on the steering wheel deflection angle and the angular transmission ratio of the inner wheel steering system, using the formula... Determine the inner wheel deflection angle; where δ0 is the inner wheel deflection angle; δ is the steering wheel deflection angle; i ω1 This refers to the angular transmission ratio of the inner wheel steering system.

[0025] Based on the steering wheel deflection angle and the angular transmission ratio of the outer wheel steering system, using the formula... Determine the outer wheel deflection angle; where δ i The outer wheel deflection angle; i ω2 This refers to the angular transmission ratio of the outer wheel steering system.

[0026] Based on the outer wheel deflection angle, using the formula Determine the turning radius; where R0 is the turning radius and L is the wheelbase;

[0027] Based on the vehicle speed and the turning radius, using the formula Determine the lateral reaction force exerted on the wheeled vehicle by the ground; where F Y The lateral reaction force exerted by the ground on the wheeled vehicle is M; the vehicle's curb weight is v; and the vehicle speed is v.

[0028] Based on the lateral reaction force from the ground and the tire lateral stiffness experienced by the wheeled vehicle, using formula F... Y / 4=kα determines the wheel slip angle; k is the tire slip stiffness; α is the wheel slip angle;

[0029] Based on the wheel slip angle, the pre-driving path of each wheel in the wheeled vehicle is determined in the three-dimensional map using the Ackerman model.

[0030] Optionally, based on the aiming time and the pre-driving path of each wheel, the aiming point of each wheel is determined, including:

[0031] Based on the vehicle speed and the turning radius, the turning angular velocity of the wheeled vehicle is determined using the formula ω=v / R0; where ω is the turning angular velocity.

[0032] The product of the aiming time and the turning angular velocity of the wheeled vehicle is determined as the aiming angle;

[0033] Select any wheel as the current wheel;

[0034] Starting from the coordinates of the current wheel at the beginning of the preview, the preview path is intercepted on the preview path of the current wheel according to the preview angle.

[0035] When the current wheel is the inner wheel, the endpoint of the aiming path is determined based on the inner wheel's deflection angle and is the aiming point of the current wheel.

[0036] When the current wheel is the outer wheel, the endpoint of the aiming path is determined based on the outer wheel's deflection angle and is the aiming point of the current wheel.

[0037] Update the current wheel and return to the step "starting from the coordinates of the current wheel at the start of the preview, intercept the preview path on the preview path of the current wheel according to the preview angle" until all wheels are traversed to obtain the preview point of each wheel.

[0038] Optionally, the elevation data of the aiming area corresponding to the aiming point of each wheel is extracted from the 3D map, including:

[0039] Select any wheel as the current wheel;

[0040] With the current wheel's aiming point as the center and the current wheel's pre-driving path direction as the major axis, construct an elliptical region as the current wheel's aiming region.

[0041] Update the current wheel and return to the step "Construct an elliptical region as the current wheel's pre-aiming point as the center and the current wheel's pre-driving path direction as the major axis" until all wheels are traversed, resulting in multiple pre-aiming regions.

[0042] Elevation data for each preview area is extracted from the 3D map.

[0043] A road surface prediction system considering steering operations of wheeled vehicles includes:

[0044] The point cloud data acquisition module is used to acquire point cloud data of the road surface in front of the wheeled vehicle when the steering wheel deflection angle is detected.

[0045] The 3D map building module is used to build a 3D map of the road surface in front of the wheels based on point cloud data of the road surface in front of the wheeled vehicle.

[0046] The vehicle speed acquisition module is used to acquire the speed of wheeled vehicles;

[0047] The pre-driving path determination module is used to determine the pre-driving path of each wheel in the wheeled vehicle based on the steering wheel deflection angle and the vehicle speed.

[0048] The aiming point determination module is used to determine the aiming point of each wheel based on the aiming time and the pre-driving path of each wheel;

[0049] The elevation data extraction module is used to extract the elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel from the three-dimensional map.

[0050] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to cause the electronic device to perform a road surface preview method that takes into account the steering operation of a wheeled vehicle.

[0051] Optionally, the memory is a readable storage medium.

[0052] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0053] This invention provides a road surface pre-aiming method, system, and device considering the steering operation of wheeled vehicles. By installing multiple sensors, it collects 3D point cloud data of the road surface in front of the left and right wheels in real time. Point cloud filtering is used to remove noise points, outliers, and holes, followed by point cloud downsampling to reduce point cloud density, decrease computational load, and ensure real-time mapping. Considering the vehicle's bumpy ride on uneven roads, point cloud registration is used to integrate point cloud data from different perspectives into a single specified coordinate system through rigid transformation, ensuring the accuracy of the real-time map. The tire elevation sequence in the terrain mesh is extracted in real time to obtain the unevenness information of the road surface in front of the vehicle, which, combined with the vehicle's current state, serves as the input to the suspension control system. Furthermore, based on the vehicle's straight-line driving condition, a pre-aiming method based on steering conditions is proposed, and a method for calculating the pre-aiming range of the lidar during steering is demonstrated. The influence of tire lateral deviation on pre-aiming during steering is also addressed, which can improve road surface pre-aiming accuracy and thus improve the control accuracy of wheeled vehicles. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 This is a flowchart of the road surface pre-aiming method considering the steering operation of wheeled vehicles in Embodiment 1 of the present invention;

[0056] Figure 2 This is a flowchart of the active suspension anti-aiming control system in Embodiment 1 of the present invention;

[0057] Figure 3 This is a schematic diagram of the vehicle front aiming process and coordinate system in Embodiment 1 of the present invention;

[0058] Figure 4This is a schematic diagram of the front and rear wheel trajectories and aiming points of a vehicle based on the Ackerman model in Embodiment 1 of the present invention.

[0059] Figure 5 This is the projection of the road surface point cloud information within the tire ground contact ellipse in the xy direction in Embodiment 1 of the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] The purpose of this invention is to provide a road surface prediction method, system, and device that takes into account the steering operation of wheeled vehicles, which can improve the road surface prediction accuracy and thus improve the control accuracy of wheeled vehicles.

[0062] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] Example 1

[0064] like Figure 1 As shown, this embodiment provides a road surface pre-aiming method considering the steering operation of a wheeled vehicle. The method is applied to a wheeled vehicle equipped with multiple solid-state LiDARs. The solid-state LiDARs are used to acquire point cloud data of the road surface in front of the wheeled vehicle.

[0065] The methods include:

[0066] Step 101: When the steering wheel deflection angle is detected, acquire point cloud data of the road surface in front of the wheeled vehicle.

[0067] Step 102: Construct a 3D map of the road surface in front of the wheel based on the point cloud data of the road surface in front of the wheel vehicle.

[0068] Step 103: Obtain the speed of the wheeled vehicle.

[0069] Step 104: Determine the pre-driving path of each wheel in the wheeled vehicle based on the steering wheel deflection angle and vehicle speed.

[0070] Step 105: Determine the aiming point of each wheel based on the aiming time and the pre-driving path of each wheel.

[0071] Step 106: Extract the elevation data of the aiming area corresponding to the aiming point of each wheel from the 3D map.

[0072] Step 107: Input the elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel as the real-time road excitation into the suspension control system of the wheeled vehicle.

[0073] Step 102 includes:

[0074] Step 102-1: Filter the point cloud data of the road surface in front of the wheeled vehicle to obtain multiple sets of filtered point cloud data.

[0075] Step 102-2: Perform downsampling processing on the multiple sets of filtered point cloud data to obtain multiple sets of downsampled point cloud data.

[0076] Step 102-3: Extract feature points from each group of downsampled point cloud data to obtain the feature points of each group of downsampled point cloud data.

[0077] Step 102-4: Based on feature points, perform coordinate registration on multiple sets of downsampled point cloud data, transform the multiple sets of downsampled point cloud data into the same coordinate system, and obtain a three-dimensional map of the road surface in front of the wheel.

[0078] Step 104 includes:

[0079] Step 104-1: Obtain the current inner wheel steering system angular transmission ratio and outer wheel steering system angular transmission ratio;

[0080] Step 104-12: Based on the steering wheel deflection angle and the angular transmission ratio of the inner wheel steering system, use the formula... Determine the inner wheel deflection angle; where δ0 is the inner wheel deflection angle; δ is the steering wheel deflection angle; i ω1 This refers to the angular transmission ratio of the inner wheel steering system.

[0081] Step 104-3: Based on the steering wheel deflection angle and the angular transmission ratio of the outer wheel steering system, use the formula... Determine the outer wheel deflection angle; where δ i The outer wheel deflection angle; i ω2 This refers to the angular transmission ratio of the outer wheel steering system.

[0082] Step 104-4: Based on the outer wheel deflection angle, use the formula Determine the turning radius; where R0 is the turning radius and L is the wheelbase;

[0083] Step 104-5: Based on the vehicle speed and the turning radius, use the formula Determine the lateral reaction force exerted on the wheeled vehicle by the ground; where F Y The lateral reaction force exerted by the ground on the wheeled vehicle is M; the vehicle's curb weight is v; and the vehicle speed is v.

[0084] Step 104-6: Based on the lateral reaction force from the ground and the tire lateral stiffness experienced by the wheeled vehicle, use formula F Y / 4=kα determines the wheel slip angle; k is the tire slip stiffness; α is the wheel slip angle;

[0085] Step 104-7: Based on the wheel slip angle, use the Ackerman model to determine the pre-driving path of each wheel in the wheeled vehicle in the three-dimensional map.

[0086] Step 105 includes:

[0087] Step 105-1: Based on the vehicle speed and turning radius, determine the turning angular velocity of the wheeled vehicle using the formula ω=v / R0. Where ω is the turning angular velocity.

[0088] Step 105-2: Determine the product of the aiming time and the turning angular velocity of the wheeled vehicle as the aiming angle.

[0089] Step 105-3: Determine any wheel as the current wheel.

[0090] Step 105-4: Starting from the coordinates of the current wheel at the beginning of the preview, extract the preview path on the preview path of the current wheel according to the preview angle.

[0091] Step 105-5: Determine the endpoint of the aiming path as the aiming point of the current wheel. Specifically, when the current wheel is the inner wheel, determine the endpoint of the aiming path as the aiming point of the current wheel based on the inner wheel's deflection angle; when the current wheel is the outer wheel, determine the endpoint of the aiming path as the aiming point of the current wheel based on the outer wheel's deflection angle.

[0092] Step 105-6: Update the current wheel and return to step 105-4 until all wheels have been traversed, obtaining the aiming point of each wheel.

[0093] Step 106 includes:

[0094] Step 106-1: Determine any wheel as the current wheel.

[0095] Step 106-2: Using the current wheel's aiming point as the center and the current wheel's pre-driving path direction as the major axis, construct an elliptical region as the current wheel's aiming region.

[0096] Step 106-3: Update the current wheel and return to step 106-2 until all wheels have been traversed, resulting in multiple pre-aiming areas.

[0097] Step 106-4: Extract elevation data for each preview area from the 3D map.

[0098] like Figure 2The basic principle of this embodiment is as follows: First, a three-dimensional map of the road surface in front of the wheels is created based on the road surface preview sensor. Then, the driving paths of the front and rear wheels are calculated based on the front wheel deflection angle when the wheeled vehicle is turning. Next, the local road surface that the front and rear wheels will contact at a certain future moment is determined. Finally, the elevation information of the local road surface is extracted and processed to provide input excitation for the suspension control system. The vehicle's state information serves as the feedback quantity for the suspension system, which is achieved through the following technical solution:

[0099] Choosing the right road surface pre-aiming sensor is crucial. Because solid-state LiDAR has a relatively small field of view, multiple LiDARs are typically used for road mapping. LiDARs can be installed in front of the left and right front wheels of the vehicle. Steering wheel angle and vehicle speed data can be obtained from the vehicle's bus system.

[0100] Step 1: Create a 3D map of the road surface in front of the wheels based on the road surface preview sensor.

[0101] The specific process is as follows: First, solid-state lidar is used to collect point cloud data and perform point cloud filtering. Then, the point cloud data is downsampled to reduce the point cloud density. Next, point cloud feature points are extracted. Finally, the point cloud data collected by the lidar in different poses is integrated into the same coordinate system through registration processing.

[0102] Step 2: Calculate the path that the front and rear wheels will travel based on the steering wheel deflection angle and vehicle speed.

[0103] The specific process is as follows: First, calculate the left and right steering wheel deflection angles based on the steering wheel deflection angle and the angular transmission ratio of the car steering system, and calculate the steering radius based on the Ackermann steering principle. Then, calculate the lateral force of the car based on the vehicle speed and the steering radius, and calculate the side slip angles of the front and rear wheels based on the tire side slip characteristics. Finally, calculate the arc path that the front and rear wheels will travel based on the actual rolling direction of each wheel.

[0104] Step 3: Calculate the path points (pre-aiming points) that the front and rear wheels will reach based on the pre-aiming time.

[0105] The specific process is as follows: First, calculate the rotational speed of the front and rear wheels based on the vehicle speed and the state of the vehicle's steering. Then, calculate the distance that the front and rear wheels will roll based on the given aiming time. Finally, determine the path point (aiming point) that the wheels will reach on the arc path of the front and rear wheels.

[0106] The road surface pre-aiming sensors should be installed so that the pre-aiming points of each wheel are within their detection range.

[0107] Step 4: Extract the elevation information of each wheel aiming point and the road surface near it, and process the data.

[0108] The specific process is as follows: First, the tire contact area is regarded as an ellipse and the major and minor axis parameters of the ellipse are determined according to the tire contact conditions. Then, a three-dimensional map of point cloud is established in the world coordinate system and converted into road surface elevation information. The road surface elevation information of all points in the tire contact ellipse at the pre-aiming point of each wheel is extracted. Then, the extracted road surface elevation information is weighted and averaged to obtain the elevation value of the pre-aiming point of each wheel, which is used as the real-time road excitation input of the corresponding suspension control system.

[0109] The implementation method is as follows:

[0110] Step 1: Create a 3D map of the road surface in front of the wheels based on the road surface preview sensor.

[0111] (1) Use solid-state lidar to collect point cloud data and perform point cloud filtering.

[0112] Solid-state lidar collects dense point cloud data, and the quality of this data significantly impacts the accuracy of real-time localization and map building. Therefore, preprocessing of this large volume of point cloud data is essential. The first step in point cloud processing is point cloud filtering, which primarily removes noise points, outliers, and holes, laying the foundation for subsequent processing.

[0113] (2) Downsample the point cloud data.

[0114] Downsampling works by dividing the point cloud within a large cube in space into many smaller cubes (voxel meshes). The centroid of each small cube is determined based on the point cloud density, and this centroid is considered the equivalent point cloud for that region. Therefore, the equivalent coordinates x, y, z of each voxel mesh are:

[0115]

[0116] N is the number of points within the voxel grid, x i y i , z i Let i be the three-dimensional coordinates of each point, i = 1, 2, ..., n.

[0117] (3) Extraction of feature points from point cloud.

[0118] It can extract the edge features of objects, which is crucial for subsequent point cloud registration. For the measured point A, the following formula applies:

[0119]

[0120] O represents the location of the lidar, and B, C, D, and E are four points surrounding the measured point A. The length of the sum of these five vectors is used to determine the characteristics of point A. If the value of the vector sum is less than a certain threshold, A is considered a planar point. Next, the angle between two planes near point A is calculated:

[0121]

[0122] In the formula These are the normal vectors of two planes near A. If the value of θ is less than a certain threshold, then θ can be determined as an edge point.

[0123] (4) Perform registration processing on the point cloud data.

[0124] Point cloud registration works by calculating coordinate transformations, which rigidly integrate point cloud data from different viewpoints into a single specified coordinate system. When the radar's pose changes, point clouds acquired at different times will not completely overlap; therefore, point cloud registration is necessary to generate a map.

[0125] The principle of point cloud coordinate transformation is as follows: for two points X and X' in different coordinate systems:

[0126] X'=R 3×3 X+T 3×1 .

[0127] R is the rotation matrix, and T is the translation matrix. If the rotation angles of the coordinate axes along the x, y, and z axes are α, β, and γ, respectively, then the rotation matrix can be expressed as:

[0128]

[0129] Further map construction is performed, selecting an octomap as needed. Octomaps are a flexible, compressible, and constantly updatable map format. Due to accumulated errors, map construction may introduce inaccuracies. Therefore, loop closure detection is introduced to eliminate these errors. When a vehicle reaches the same position as before, loop closure detection is triggered, optimizing the pose and continuously improving map accuracy.

[0130] Step 2: Calculate the path that the front and rear wheels will travel based on the steering wheel deflection angle and vehicle speed.

[0131] The inner and outer steering wheel deflection angles are calculated based on the steering wheel angle and the angular transmission ratio of the steering system.

[0132] When the vehicle turns, the steering wheel deflection angle δ can be obtained from the bus, and the angular transmission ratio of the inner and outer wheel steering systems is i. ω1 i ω2 The deflection angles δ0 and δi of the inner and outer wheels of the vehicle can be calculated:

[0133]

[0134]

[0135] (2) Calculate the steering radius based on the Ackermann steering principle.

[0136] The turning radius R0 can be calculated using the following formula:

[0137]

[0138] (3) Calculate the lateral reaction force F of the vehicle on the ground based on the vehicle speed v and the turning radius R0. Y .

[0139] Since the lateral reaction force from the ground acting on the car is equal to the centrifugal force generated during turning, the lateral reaction force F acting on the car can be calculated using the following formula. Y :

[0140]

[0141] Calculate the slip angles of the front and rear wheels based on the tire slip characteristics.

[0142] The lateral reaction force from the ground is the resultant force of the lateral reaction force exerted by the ground on each tire. For a four-wheeled vehicle, the lateral reaction force from the ground on a single tire can be approximated as 1 / 4 of the lateral reaction force from the ground acting on the vehicle, i.e., F. Y =1 / 4F Y The rolling tire will generate a sideslip angle α under this force. When the lateral reaction force from the ground on the car is small, it can be approximated by the following linear relationship:

[0143] F Y =kα

[0144] In the formula, K is the tire lateral stiffness, which is generally taken in the range of -80000 to -28000.

[0145] (5) Calculate the actual circular arc trajectory of each wheel.

[0146] When the sideslip angle is less than a certain value, the actual turning process of the vehicle should satisfy the Ackermann model, with the instantaneous steering center of each wheel at point O. The circular trajectory of the wheels is shown in the appendix. Figure 4 .

[0147] Step 3: Calculate the path points (pre-aiming points) that the front and rear wheels will reach based on the pre-aiming time. A diagram illustrating the vehicle's front-end pre-aiming process and coordinate system is shown below. Figure 3 As shown.

[0148] (1) Calculate the turning angular velocity of the vehicle based on the vehicle speed and turning radius.

[0149] When the vehicle speed is v and the turning radius is R0, the vehicle's turning angular velocity ω = v / R0.

[0150] (2) Calculate the distance that each of the front and rear wheels will roll based on the given aiming time.

[0151] The reaction time allowed for the suspension is t. s When the car's turning angular velocity is ω, the angle of rotation of the wheel relative to the center of rotation O is ωt. s The turning trajectory is the range defined by the angle of the turn.

[0152] (3) Determine the path points (pre-aiming points) that the front and rear wheels will reach.

[0153] Let the four wheel aiming points be D. fl D fr D rl D rr .

[0154] Left front wheel, right front wheel D fl With D fr The x and y coordinates (with the forward direction as x) relative to the origin of the coordinate system, with the corresponding wheel contact point as the origin, are as follows:

[0155]

[0156]

[0157]

[0158]

[0159] Left rear wheel, right rear wheel D rl With D rr The x and y coordinates relative to the origin of the coordinate system, with the corresponding wheel contact point as the reference point, are as follows:

[0160]

[0161]

[0162]

[0163]

[0164] δ0 and δi are the deflection angles of the inner and outer wheels of the vehicle, ω is the turning angular velocity, and t is the turning angle. s This refers to the suspension anti-aiming time.

[0165] Verify that the aiming point is within the scanning range of the lidar.

[0166] Based on geometric relationships, the turning angular velocity and the constraint relationship between the front wheel steering angle and the radar detection angle θ1 can be calculated.

[0167] The radar located on the left side has the following functions for the left front wheel and left rear wheel respectively:

[0168]

[0169]

[0170] The radar located on the right side has separate functions for the right front wheel and the right rear wheel.

[0171]

[0172]

[0173] When a car turns right, the radar on the right front wheel is prone to blind spots while anticipating the road surface that the right rear wheel will pass. Similarly, when turning left, the radar on the left side is also prone to blind spots. This can be addressed by adding steering assist radars on both sides to compensate for the blind spots during turns. When the steering wheel angle is detected, the side radars automatically begin detection.

[0174] Step 4: Extract the elevation information of each wheel aiming point and the road surface near it, and process the data.

[0175] (1) Determine the major and minor axis parameters of the tire contact ellipse based on the tire contact conditions.

[0176] The tire contact patch is related to various factors such as tire pressure and load. For ordinary passenger cars, the ratio of the major and minor axes of the contact patch ellipse is approximately 0.9 to 1.05. According to research, when the tire pressure is 0.8 MPa and the load is 10 kN, the tire contact patch is approximately 1.25 × 10³ mm. 2 .

[0177] A 3D map of point clouds built in the world coordinate system is converted into road surface elevation information.

[0178] Using the vehicle's aiming point as the center of an ellipse, point cloud information within the tire contact ellipse area is extracted, and the processed point cloud data is projected, as shown in the attached figure. Figure 5 As shown, the road surface elevation sequence is obtained, and the calculation formula is:

[0179]

[0180] Where p is the road surface elevation coordinate obtained by projection, and r SP This refers to the road surface data obtained from LiDAR scanning in front of the vehicle. and r SMIt is the rotation matrix and translation vector from the lidar coordinate system to the world coordinate system.

[0181] A weighted average is performed to obtain the elevation value of the aiming point of each wheel.

[0182] Elevation values ​​(h) of all points within the ellipse j The weighted average of (j = 1, 2, ..., n) is used as the tire contact point (h) i The elevation value is calculated using the following formula:

[0183]

[0184] In the formula, w j The weight is determined by the distance from the grounding point; the closer the distance is to the grounding point, the greater the weight.

[0185] The following detailed explanation of this embodiment will be provided with specific examples.

[0186] Step 1: At the start of each pre-aiming cycle, after the solid-state LiDAR receives the point cloud information, it retains the xyz values ​​of the point cloud data. First, it surrounds all the point cloud data with a large cube and calculates the maximum value of all point clouds in each dimension, which serves as the boundary of the large cube. Then, it customizes the size r of the voxel mesh, divides the large cube into many small cubes, and calculates the centroid of the point cloud of the small cube, which serves as the new point cloud.

[0187] Step 2: Use IMU sensors to sense changes in vehicle pose, and then perform pose transformation between point clouds of different frames. Combine the data scanned in the first k frames according to the pose change relationship to form a local map to obtain more accurate road information.

[0188] Step 3: The pre-aiming process involves four coordinate systems: world coordinate system I, vehicle coordinate system B, radar coordinate system S, and elevation map coordinate system M, as shown in the attached diagram. Figure 3 As shown.

[0189] According to the IMU, the translation r of the vehicle coordinate system relative to the world coordinate system can be obtained. IB and rotation (Including pitch, yaw, roll), as well as the radar coordinate system, elevation map coordinate system, and translation r relative to the vehicle coordinate system. BS r BM , This allows us to obtain the pose r of the elevation map coordinate system relative to the radar coordinate system. SM ,

[0190] Step 4: The radar coordinate system is approximately 0.5m above the ground. The elevation map coordinate system is located at the wheel contact point at the current moment. r can be obtained in step 2. SM , P is the identity matrix. The coordinates r of the point cloud detected by the radar are transformed using coordinate transformation. SP Mapping to the elevation map coordinate system yields p:

[0191]

[0192] Step 5: Calculate using the formula above. When the vehicle turns at 15 km / h with a preview time of 0.5 seconds, the inner and outer wheel turning angles δ... i δ0 is 20° and 17° respectively, and the wheelbase is l. w =1.3m, wheelbase L = 2.5m, the turning radius R0 = 6.87m, the turning angular velocity is approximately ω = 0.61rad / s, and the xy coordinates of the aiming point relative to the contact points of each wheel can be calculated as follows:

[0193] left front wheel right front wheel

[0194] Left rear wheel Right rear wheel

[0195] Step 6: Extract elevation information within the tire contact ellipse area based on the pre-aiming point, and perform weighted calculation of the elevation value. The tire contact area is approximately 1.25 × 10³ mm. 2 The major and minor axes are calculated to be approximately 21 mm and 19 mm, respectively. Based on the point cloud information in the elevation map coordinate system, the z coordinates corresponding to all x and y points falling within the ellipse can be obtained, which are used as the elevation value h of that point.

[0196] After weighting

[0197]

[0198] The elevation of the aiming point is obtained and used as the control input for the suspension.

[0199] Example 2

[0200] In order to execute the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a road surface pre-aiming system considering the steering operation of wheeled vehicles is provided below, including:

[0201] The point cloud data acquisition module is used to acquire point cloud data of the road surface in front of the wheeled vehicle when the steering wheel deflection angle is detected.

[0202] The 3D map building module is used to build a 3D map of the road surface in front of the wheels based on point cloud data of the road surface in front of wheeled vehicles.

[0203] The vehicle speed acquisition module is used to acquire the speed of wheeled vehicles.

[0204] The pre-driving path determination module is used to determine the pre-driving path of each wheel in a wheeled vehicle based on the steering wheel deflection angle and vehicle speed.

[0205] The aiming point determination module is used to determine the aiming point of each wheel based on the aiming time and the pre-driving path of each wheel.

[0206] The elevation data extraction module is used to extract the elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel from the 3D map.

[0207] Example 3

[0208] This embodiment provides an electronic device, including a memory and a processor. The memory stores a computer program, and the processor runs the computer program to enable the electronic device to perform a road surface pre-aiming method considering the steering operation of a wheeled vehicle as described in Embodiment 1.

[0209] The memory is a readable storage medium.

[0210] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0211] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of these embodiments are merely for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A road surface pre-aiming method considering the steering operation of wheeled vehicles, characterized in that, The method is applied to a wheeled vehicle equipped with multiple solid-state lidar sensors. The solid-state lidar is used to acquire point cloud data of the road surface in front of the wheeled vehicle; The method includes: When the steering wheel deflection angle is detected, point cloud data of the road surface in front of the wheeled vehicle is acquired; A 3D map of the road surface in front of the wheels is constructed based on point cloud data of the road surface in front of the wheeled vehicle. Obtain the speed of wheeled vehicles; Based on the steering wheel deflection angle and the vehicle speed, determine the pre-driving path of each wheel in the wheeled vehicle; Based on the aiming time and the pre-driving path of each wheel, determine the aiming point of each wheel, including: based on vehicle speed and turning radius, using the formula... Determine the turning angular velocity of the wheeled vehicle; where, The turning angular velocity is used as the reference value. The product of the aiming time and the turning angular velocity of the wheeled vehicle is determined as the aiming angle. Any wheel is identified as the current wheel. Starting from the coordinates of the current wheel at the start of aiming, the aiming path is intercepted on the pre-driving path of the current wheel according to the aiming angle. When the current wheel is the inner wheel, the end point of the aiming path is determined as the aiming point of the current wheel according to the inner wheel deflection angle. When the current wheel is the outer wheel, the end point of the aiming path is determined as the aiming point of the current wheel according to the outer wheel deflection angle. The current wheel is updated and the process returns to the step "starting from the coordinates of the current wheel at the start of aiming, the aiming path is intercepted on the pre-driving path of the current wheel according to the aiming angle" until all wheels are traversed to obtain the aiming point of each wheel. The process of extracting elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel from the 3D map includes: determining any wheel as the current wheel; constructing an elliptical region as the pre-aiming area of ​​the current wheel with the pre-aiming point of the current wheel as the center and the direction of the pre-driving path of the current wheel as the major axis; updating the current wheel and returning to the step "constructing an elliptical region as the pre-aiming area of ​​the current wheel with the pre-aiming point of the current wheel as the center and the direction of the pre-driving path of the current wheel as the major axis" until all wheels are traversed to obtain multiple pre-aiming areas; and extracting elevation data of each pre-aiming area from the 3D map.

2. The road surface pre-aiming method considering the steering operation of wheeled vehicles according to claim 1, characterized in that, After extracting the elevation data of the aiming area corresponding to the aiming point of each wheel from the 3D map, the method further includes: The elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel is used as the real-time input of the road excitation and input into the suspension control system of the wheeled vehicle.

3. The road surface pre-aiming method considering the steering operation of wheeled vehicles according to claim 1, characterized in that, The construction of a 3D map of the road surface in front of the wheels based on point cloud data of the road surface in front of the wheeled vehicle includes: The point cloud data of the road surface in front of the wheeled vehicle were filtered to obtain multiple sets of filtered point cloud data. The point cloud data after multiple filtering processes are downsampled to obtain multiple sets of downsampled point cloud data. Feature points were extracted from each group of downsampled point cloud data to obtain the feature points of each group of downsampled point cloud data. Based on the feature points, coordinate registration is performed on multiple sets of downsampled point cloud data, and the multiple sets of downsampled point cloud data are transformed into the same coordinate system to obtain a three-dimensional map of the road surface in front of the wheel.

4. The road surface pre-aiming method considering the steering operation of wheeled vehicles according to claim 1, characterized in that, Based on the steering wheel deflection angle and the vehicle speed, the pre-travel path of each wheel in the wheeled vehicle is determined, including: Obtain the current steering ratio of the inner wheel and the steering ratio of the outer wheel. Based on the steering wheel deflection angle and the angular transmission ratio of the inner wheel steering system, using the formula... Determine the inner wheel deflection angle; where, The inner wheel deflection angle; This refers to the steering wheel deflection angle; This refers to the angular transmission ratio of the inner wheel steering system. Based on the steering wheel deflection angle and the angular transmission ratio of the outer wheel steering system, using the formula... Determine the outer wheel deflection angle; where, This refers to the outer wheel deflection angle; This refers to the angular transmission ratio of the outer wheel steering system. Based on the outer wheel deflection angle, using the formula Determine the turning radius; where, The turning radius; This refers to the wheelbase; Based on the vehicle speed and the turning radius, using the formula Determine the lateral reaction forces exerted on the wheeled vehicle by the ground; among which, This refers to the lateral reaction force exerted by the ground on a wheeled vehicle. For vehicle curb weight; For vehicle speed; Based on the lateral reaction force from the ground and the tire lateral stiffness experienced by the wheeled vehicle, using the formula... Determine the wheel slip angle; This refers to the tire's lateral stiffness. This refers to the wheel slip angle; Based on the wheel slip angle, the pre-driving path of each wheel in the wheeled vehicle is determined in the three-dimensional map using the Ackerman model.

5. A road surface prediction system considering steering operations of wheeled vehicles, characterized in that, The system is used to perform the road surface pre-aiming method considering wheeled vehicle steering operations as described in any one of claims 1-4, the system comprising: The point cloud data acquisition module is used to acquire point cloud data of the road surface in front of the wheeled vehicle when the steering wheel deflection angle is detected. The 3D map building module is used to build a 3D map of the road surface in front of the wheels based on point cloud data of the road surface in front of the wheeled vehicle. The vehicle speed acquisition module is used to acquire the speed of wheeled vehicles; The pre-driving path determination module is used to determine the pre-driving path of each wheel in the wheeled vehicle based on the steering wheel deflection angle and the vehicle speed. The aiming point determination module is used to determine the aiming point of each wheel based on the aiming time and the pre-driving path of each wheel; The elevation data extraction module is used to extract the elevation data of the pre-aiming area corresponding to the pre-aiming point of each wheel from the three-dimensional map.

6. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform a road surface pre-aiming method considering steering operations of a wheeled vehicle, as described in any one of claims 1 to 4.

7. An electronic device according to claim 6, characterized in that, The memory is a readable storage medium.

Images