A vehicle automatic parking method, electronic device and storage medium

Abstract

This application provides a method, electronic device, and storage medium for automatic vehicle parking, applicable to the fields of autonomous driving, driverless driving, and unmanned vehicle technology. The method includes: independently performing deceleration acquisition operations under different scenario dimensions using a look-ahead path curvature interval hysteresis mechanism, a vehicle roll angle lateral slope interval hysteresis mechanism, and a road friction hysteresis mechanism; performing a deceleration arbitration operation on the decelerations under different scenario dimensions to obtain the desired deceleration for the scenario; using the road friction coefficient and the vehicle's current lateral acceleration to apply a friction circle constraint in the longitudinal direction to obtain the maximum longitudinal deceleration; performing a deceleration arbitration operation again on the desired deceleration and the maximum longitudinal deceleration to obtain the target braking command; locking the steering angle command at the moment of vehicle malfunction; and using the target braking command to perform parking deceleration control on the vehicle until the vehicle's real-time speed meets a preset parking speed threshold, thus completing the automatic parking of the vehicle in emergency stop mode.

Description

Technical Field

[0001] This application relates to the fields of autonomous driving, driverless driving and driverless vehicle technology, and specifically to a method for automatic vehicle parking, electronic equipment and storage medium. Background Technology

[0002] In autonomous driving systems, when the fault diagnosis module detects sensor failure, control anomalies, or a deterioration in system health, it needs to trigger an emergency stop and degrade command to bring the vehicle to a safe stop. However, most existing fault-stop strategies use fixed deceleration to perform braking, which presents the following problems under complex road conditions: risk of sudden braking on sharp curves; risk of slippage on low-adhesion surfaces; risk of rollover on steep cross slopes; and scene transition jitter issues. Summary of the Invention

[0003] In view of the above problems, embodiments of this application provide a method for automatic parking of a vehicle, an electronic device, and a storage medium.

[0004] The first aspect of this application provides a method for automatic vehicle parking, comprising: when the received parking downgrade command is an emergency stop mode, independently performing deceleration acquisition operations under different scenario dimensions using a look-ahead path curvature interval hysteresis mechanism, a vehicle roll angle lateral slope interval hysteresis mechanism, and a road surface friction hysteresis mechanism, and performing a deceleration arbitration operation on the deceleration under different scenario dimensions to obtain the desired deceleration for the scenario; using the road surface friction coefficient and the vehicle's current lateral acceleration to perform friction circle constraint processing on the vehicle's longitudinal control to obtain the maximum longitudinal deceleration; performing a deceleration arbitration operation again on the desired deceleration for the scenario and the maximum longitudinal deceleration to obtain a target braking command; performing lateral control on the vehicle by locking the steering angle command at the moment the vehicle malfunctions, and performing parking deceleration control on the vehicle using the target braking command until the vehicle's real-time speed meets a preset parking speed threshold, thereby completing the automatic parking of the vehicle in emergency stop mode.

[0005] According to embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scenario dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road friction hysteresis mechanism includes: determining the look-ahead arc length window based on the vehicle's current speed, and extracting curvature data falling within the look-ahead arc length window from the vehicle's planned path; performing sorting and averaging operations on the extracted curvature data to obtain the curvature mean, and using the curvature mean to determine the look-ahead equivalent radius; mapping the look-ahead equivalent radius to a multi-level curvature interval, and using the curvature interval hysteresis to control the state transition of the multi-level curvature interval to obtain a multi-level curvature interval with no jitter in the boundary switching region; and mapping the current state information of the multi-level curvature interval with no jitter in the boundary switching region to obtain the deceleration under the look-ahead path curvature dimension scenario.

[0006] According to an embodiment of this application, the above-mentioned sorting and averaging operations on the extracted curvature data to obtain the average curvature value include: performing a descending sorting operation on the absolute values ​​of the extracted curvature data to obtain a descending sorting result; selecting multiple top-ranking absolute curvature values ​​from the descending sorting result, and performing an averaging operation on the multiple absolute curvature values ​​to obtain the average curvature value.

[0007] According to an embodiment of this application, the above-mentioned control of the state transition of multi-level curvature intervals using curvature interval hysteresis to obtain multi-level curvature intervals with no jitter in the boundary switching area includes: comparing the curvature interval hysteresis, the curvature threshold of gentle curves and the curvature threshold of sharp curves with the curvature of straight road area, curvature of gentle curve area and curvature of sharp curve area in multi-level curvature intervals respectively to obtain curvature comparison results; and completing the state transition between straight road state, gentle curve state and sharp curve state in multi-level curvature intervals according to the curvature comparison results to obtain multi-level curvature intervals with no jitter in the boundary switching area.

[0008] According to the embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scene dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road friction hysteresis mechanism further includes: determining multi-level lateral slope intervals using the vehicle roll angle as a lateral slope estimate; comparing the lateral slope interval hysteresis, gentle slope threshold, and steep slope threshold with the real-time vehicle roll angles in the flat road area, gentle slope area, and steep slope area of ​​the obtained multi-level lateral slope intervals to obtain lateral slope comparison results; completing the state transition between the flat road state, gentle slope state, and steep slope state in the multi-level lateral slope intervals based on the lateral slope comparison results to obtain the lateral slope interval with no jitter in the boundary switching area; and mapping the current state information of the lateral slope interval with no jitter in the boundary switching area to obtain the deceleration under the lateral slope dimension scene.

[0009] According to the embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scenario dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road surface friction hysteresis mechanism further includes: comparing the road surface friction hysteresis amount and the preset friction threshold with the current road surface friction coefficient to obtain the road surface friction comparison result; determining the current road surface state information based on the road surface friction comparison result, and using the current road surface state information to map the deceleration under the road surface friction dimension scenario.

[0010] According to an embodiment of this application, the above-mentioned friction circle constraint processing for vehicle longitudinal control using the road surface friction coefficient and the vehicle's current lateral acceleration to obtain the maximum longitudinal deceleration includes: determining the maximum longitudinal deceleration of the vehicle in the longitudinal direction using a friction circle dynamics model under the current road surface friction coefficient and the vehicle's current lateral acceleration; performing adhesion constraint processing on the desired deceleration of the scenario to calculate the safe limit of longitudinal deceleration allowed by the tire grip force.

[0011] According to an embodiment of this application, the above-mentioned automatic vehicle parking method further includes: when the received parking downgrade instruction is a slow stop mode representing a non-emergency fault scenario, performing parking deceleration control on the vehicle according to a preset deceleration rate until the real-time vehicle speed meets a preset parking speed threshold, thereby completing the automatic parking of the vehicle in the slow stop mode.

[0012] A second aspect of this application provides an automatic vehicle parking device, comprising: a multi-dimensional scene perception module, used to independently perform deceleration acquisition operations under different scene dimensions using a forward path curvature interval hysteresis mechanism, a vehicle roll angle lateral slope interval hysteresis mechanism, and a road surface friction hysteresis mechanism when the received parking downgrade command is an emergency stop mode, and to perform a deceleration arbitration operation on the deceleration under different scene dimensions to obtain the desired deceleration of the scene; a friction circle constraint module, used to perform friction circle constraint processing on the longitudinal control of the vehicle using the road surface friction coefficient and the current lateral acceleration of the vehicle to obtain the maximum longitudinal deceleration; a deceleration arbitration module, used to perform a deceleration arbitration operation again on the desired deceleration of the scene and the maximum longitudinal deceleration to obtain the target braking command; and a parking deceleration module, used to perform lateral control of the vehicle by locking the steering angle command at the time of the vehicle malfunction, and to perform parking deceleration control on the vehicle using the target braking command until the real-time vehicle speed meets a preset parking speed threshold, thereby completing the automatic parking of the vehicle in the emergency stop mode.

[0013] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0014] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, wherein the computer program or instructions, when executed by a processor, implement the steps of the above-described method.

[0015] A fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method.

[0016] The automatic vehicle parking method provided in this application can achieve safety on high-curvature curves by automatically reducing deceleration to a gentle deceleration during sharp curves, and effectively preventing sideslip or rollover due to excessive braking force, in conjunction with friction circle constraints. It can also achieve adaptability to low-friction surfaces by automatically switching to a conservative braking strategy on low-friction surfaces such as rain and snow, and further prevent wheel lock-up through friction circle constraints. Furthermore, it can achieve high-speed, steep-slope protection by automatically limiting deceleration when driving at high speeds on steep slopes to avoid rollover risks caused by lateral shift of the center of gravity. Finally, it can achieve smooth braking by eliminating hysteresis state machine scenarios. The deceleration transition during switching is abrupt, and the braking process is smooth and shock-free; it has forward-looking safety, adjusting the braking strategy before the curve arrives based on the peak curvature analysis of the path ahead, avoiding lag; it is highly versatile, and the automatic vehicle parking method provided in this application is applicable to various commercial vehicles (e.g., tractor trucks, dump trucks, buses, etc.), and the parameters can be flexibly adapted to different vehicle models through configuration files; in addition, the automatic vehicle parking method provided in this application has the flexibility of multi-level degradation, and the dual modes of emergency stop and slow stop provide elasticity for the system safety architecture, and the optimal degradation strategy can be matched according to the severity of the fault. Attached Figure Description

[0017] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0018] Figure 1 The illustration shows an application scenario diagram of automatic vehicle parking according to an embodiment of this application.

[0019] Figure 2 A flowchart illustrating an automatic vehicle parking method according to an embodiment of this application is shown schematically.

[0020] Figure 3 The flowchart illustrates an arbitration scheme for automatic driving fault-safe parking deceleration based on multi-dimensional scene recognition and friction circle constraints according to an embodiment of this application.

[0021] Figure 4 A schematic block diagram of a vehicle automatic parking device according to an embodiment of this application is shown.

[0022] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing an automatic vehicle parking method according to an embodiment of this application. Detailed Implementation

[0023] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0024] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0025] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0026] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0027] Most existing fault parking strategies use fixed deceleration to perform braking. Such schemes have the following problems under complex road conditions: (1) Risk of sudden braking on curves with large curvature: When performing large deceleration braking in small radius curves, the superposition of tire lateral force and longitudinal braking force can easily exceed the tire friction limit, causing the vehicle to lose control and skid or even roll over, which is especially dangerous for commercial heavy trucks. (2) Risk of slipping on low-friction road surfaces: Under low-friction road surface conditions such as rain, snow, and ice, performing the same braking intensity as on normal road surfaces will cause the wheels to lock up or skid significantly, and the vehicle will deviate from the intended driving path. (3) Risk of rollover on steep cross slopes: On road sections with large lateral slopes (such as the side slope of mountain roads and the superelevation section of ramps), performing sudden braking at high speed will cause the center of gravity load to shift sharply to the lower side. Combined with the cross slope component force, it is very easy to trigger a rollover. (4) Scene switching jitter problem: When the vehicle is in the boundary state between two scenes (such as the transition from a straight road to a curve), the scene judgment may switch frequently between the two states, causing the target deceleration to oscillate violently, affecting the ride comfort and braking system life.

[0028] To address at least one of the problems in the existing technology, this application provides an automatic parking method that uses a deceleration arbitration approach that combines multi-dimensional scene perception with friction circle constraints to enable safe automatic parking of vehicles in emergency malfunction situations. This allows autonomous vehicles to adaptively adjust braking intensity based on the current road environment when stopping due to a malfunction, achieving safe parking with optimal deceleration in high-curvature curves, steep cross slopes, low-adhesion surfaces, and normal road conditions, while balancing parking efficiency and vehicle stability.

[0029] Figure 1 The illustration shows an application scenario diagram of automatic vehicle parking according to an embodiment of this application.

[0030] like Figure 1 As shown, application scenario 100 according to an embodiment of this application may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables. For example, a user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send information, etc.

[0031] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be electronic devices such as smartphones, wearable devices, personal computers, intelligent voice interaction devices, smart home appliances, intelligent vehicles, in-vehicle terminals, aircraft, unmanned vending terminals, and extended reality devices. Extended reality devices can include virtual reality devices, augmented reality devices, and mixed reality devices. A client application for the target application can be installed and run on the terminal devices. This target application can include, but is not limited to, financial transaction applications, payment applications, shopping applications, web browser applications, search applications, instant messaging tools, email clients, and social media platform software (these are just examples). Furthermore, this application embodiment does not limit the form of the target application, and it can include, but is not limited to, applications, mini-programs, etc., installed on the terminal devices, and can also be in the form of web pages.

[0032] Server 105 can be a server providing various services, such as a backend management server supporting websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process received user requests and other data, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services such as cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks, and basic cloud computing services such as big data. The server can be the backend server of the aforementioned target application, used to provide backend services to the clients of the target application.

[0033] It should be noted that the automatic vehicle parking method provided in this application embodiment can generally be executed by server 105 and / or terminal devices 101-103. Accordingly, the automatic vehicle parking device provided in this application embodiment can generally be installed in server 105 and / or terminal devices 101-103.

[0034] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0035] Figure 2 A flowchart illustrating an automatic vehicle parking method according to an embodiment of this application is shown schematically.

[0036] like Figure 2 As shown, the automatic parking method 200 for vehicles according to an embodiment of this application may include steps S210 to S240.

[0037] In step S210, when the received parking downgrade command is an emergency stop mode, the deceleration acquisition operation under different scenario dimensions is independently performed using the forward path curvature interval hysteresis mechanism, the vehicle body roll angle lateral slope interval hysteresis mechanism, and the road surface friction hysteresis mechanism, and the deceleration under different scenario dimensions is then used to perform a deceleration arbitration operation to obtain the desired deceleration for the scenario.

[0038] Before executing step S210, a parking downgrade command issued by the vehicle fault module is received. The parking level is identified by parsing the parking downgrade command. For example, the parking level is emergency stop mode, which means that the vehicle encounters an emergency (such as a vehicle malfunction or an emergency road condition) and needs to brake quickly; or the parking level is slow stop mode, which means that the vehicle can stop slowly according to a preset deceleration, which is suitable for non-emergency situations.

[0039] Step S210 performs multi-dimensional scene classification based on the look-ahead path curvature interval hysteresis mechanism, the vehicle body roll angle lateral slope interval hysteresis mechanism, and the road surface friction hysteresis mechanism, and calls each scene classifier to update with an independent hysteresis state machine to eliminate the switching jitter between different boundaries in different dimensional scenes.

[0040] Based on the scene classification results, the recommended deceleration level for each dimension is independently mapped (i.e., based on the state interval corresponding to the current scene, the deceleration level under that state interval is obtained), and the minimum absolute value is taken as the expected deceleration of the scene.

[0041] In step S220, friction circle constraint processing is performed on the longitudinal control of the vehicle using the road friction coefficient and the current lateral acceleration of the vehicle to obtain the maximum longitudinal deceleration.

[0042] Determine the longitudinal deceleration limit of the friction circle under the current lateral acceleration: The friction circle constraint essentially limits the resultant force of the tire's longitudinal braking force / longitudinal acceleration and lateral steering force / lateral acceleration, ensuring that the total grip force cannot exceed the maximum adhesion limit of the road surface. Lateral acceleration represents the grip force resources currently occupied by the vehicle, and the remaining grip force can only be allocated to longitudinal braking; therefore, the constraint directly acts on the longitudinal braking deceleration (desired braking intensity).

[0043] In step S230, the desired deceleration and the maximum longitudinal deceleration of the scene are again subjected to deceleration arbitration operation to obtain the target braking command.

[0044] Arbitrate the expected deceleration of the scene with the friction circle limit, and take the smaller absolute value as the final braking command (i.e. the target braking command).

[0045] In step S240, the vehicle is laterally controlled by locking the steering angle command at the moment the vehicle malfunctions, and the vehicle is stopped and decelerated by using the target braking command until the real-time vehicle speed meets the preset stopping speed threshold, thus completing the automatic stopping of the vehicle in emergency stop mode.

[0046] Lateral control maintains the current steering wheel angle, preventing steering and braking from coupling. Real-time vehicle speed monitoring outputs a parking completion indicator once a preset parking speed threshold is reached.

[0047] The automatic vehicle parking method provided in this application can achieve safety on high-curvature curves by automatically reducing deceleration to a gentle deceleration during sharp curves, and effectively preventing sideslip or rollover due to excessive braking force, in conjunction with friction circle constraints. It can also achieve adaptability to low-friction surfaces by automatically switching to a conservative braking strategy on low-friction surfaces such as rain and snow, and further prevent wheel lock-up through friction circle constraints. Furthermore, it can achieve high-speed, steep-slope protection by automatically limiting deceleration when driving at high speeds on steep slopes to avoid rollover risks caused by lateral shift of the center of gravity. Finally, it can achieve smooth braking by eliminating hysteresis state machine scenarios. The deceleration transition during switching is abrupt, and the braking process is smooth and shock-free; it has forward-looking safety, adjusting the braking strategy before the curve arrives based on the peak curvature analysis of the path ahead, avoiding lag; it is highly versatile, and the automatic vehicle parking method provided in this application is applicable to various commercial vehicles (e.g., tractor trucks, dump trucks, buses, etc.), and the parameters can be flexibly adapted to different vehicle models through configuration files; in addition, the automatic vehicle parking method provided in this application has the flexibility of multi-level degradation, and the dual modes of emergency stop and slow stop provide elasticity for the system safety architecture, and the optimal degradation strategy can be matched according to the severity of the fault.

[0048] According to embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scenario dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road friction hysteresis mechanism includes: determining the look-ahead arc length window based on the vehicle's current speed, and extracting curvature data falling within the look-ahead arc length window from the vehicle's planned path; performing sorting and averaging operations on the extracted curvature data to obtain the curvature mean, and using the curvature mean to determine the look-ahead equivalent radius; mapping the look-ahead equivalent radius to a multi-level curvature interval, and using the curvature interval hysteresis to control the state transition of the multi-level curvature interval to obtain a multi-level curvature interval with no jitter in the boundary switching region; and mapping the current state information of the multi-level curvature interval with no jitter in the boundary switching region to obtain the deceleration under the look-ahead path curvature dimension scenario.

[0049] According to an embodiment of this application, the above-mentioned sorting and averaging operations on the extracted curvature data to obtain the average curvature value include: performing a descending sorting operation on the absolute values ​​of the extracted curvature data to obtain a descending sorting result; selecting multiple top-ranking absolute curvature values ​​from the descending sorting result, and performing an averaging operation on the multiple absolute curvature values ​​to obtain the average curvature value.

[0050] According to an embodiment of this application, the above-mentioned control of the state transition of multi-level curvature intervals using curvature interval hysteresis to obtain multi-level curvature intervals with no jitter in the boundary switching area includes: comparing the curvature interval hysteresis, the curvature threshold of gentle curves and the curvature threshold of sharp curves with the curvature of straight road area, curvature of gentle curve area and curvature of sharp curve area in multi-level curvature intervals respectively to obtain curvature comparison results; and completing the state transition between straight road state, gentle curve state and sharp curve state in multi-level curvature intervals according to the curvature comparison results to obtain multi-level curvature intervals with no jitter in the boundary switching area.

[0051] In one specific implementation, this application does not use the instantaneous curvature of the current position, but instead uses the peak curvature statistical method within the look-ahead preview window to construct a look-ahead path curvature analysis mechanism.

[0052] Assuming the current vehicle speed is (m / s), the look-ahead arc length window is defined as shown in formulas (1) and (2):

[0053] (1),

[0054] (2).

[0055] in and Configurable look-ahead time start and end values ​​(optionally, , ).

[0056] In planning the path In the middle, collect all that meet the requirements. absolute value of curvature Sort them in descending order and take the first few. The average of the peak values ​​is calculated as shown in formula (3):

[0057] (3).

[0058] in This represents the first element after absolute curvature is sorted in descending order within the look-ahead window. One value, Configurable peak sample count (preferred) ).

[0059] The representative curvature is converted into an equivalent radius, as shown in formula (4):

[0060] (4).

[0061] The technical effects of this mechanism are: (1) The forward window adapts to the vehicle speed, allowing for a wider view at high speeds and focusing on nearby areas at low speeds. (2) Peak values ​​are used instead of average values ​​to ensure that sharp bends that occur briefly ahead are not missed. (3) Multi-point averaging eliminates the influence of a single noise point on the results.

[0062] In the curvature interval hysteresis state machine, the look-ahead equivalent radius is mapped to a three-level curvature interval, and a hysteresis band is used to prevent boundary jitter.

[0063] Define three curvature intervals:

[0064] Straight road area: ;

[0065] Gentle curve area: ;

[0066] Sharp bend area: .

[0067] The state transition rule has curvature interval hysteresis. As shown in formulas (5) and (6):

[0068] (5),

[0069] (6).

[0070] in , and Configurable threshold (preferred) , , ).

[0071] The mechanism of adaptive scaling of the look-ahead window in the above embodiments of this application does not use a fixed number of meters as the look-ahead range, but rather obtains the look-ahead arc length window based on vehicle speed and a configurable look-ahead time. At high speeds, the travel distance corresponding to the same amount of time is longer, and the look-ahead window automatically expands; at low speeds, the look-ahead window automatically contracts, focusing on nearby roads. This solves the problems of limited visibility at high speeds and redundant previewing at low speeds when the look-ahead distance is fixed.

[0072] The embodiments of this application, after obtaining the equivalent curve radius through look-ahead calculation, do not directly map the original value to the deceleration level. Instead, they feed the value into an independent hysteresis state machine with a hysteresis band to classify the curve scene and output discrete states: straight road state, gentle curve state, and sharp curve state. Interval division thresholds are set for straight road, gentle curve, and sharp curve, and additional curvature hysteresis offsets are configured as buffer zones. The entry and exit thresholds for state switching are different: jumping from a straight road to a sharp curve and reverting from a sharp curve to a straight road use two different decision boundaries, forming a buffer "hysteresis band." Only when the change in the equivalent curve radius exceeds the hysteresis band is the state machine allowed to switch the curve classification state. This solves the problem in related technical solutions where, when there is small path noise and the equivalent curve radius fluctuates slightly near the classification threshold, the straight road-gentle curve-sharp curve state is not repeatedly switched, eliminating the ping-pong jitter at the boundary.

[0073] The key timing aspect of this application is that it first outputs a stable discrete scene state through a hysteresis state machine, and then uses the stable state to look up a table to map the recommended deceleration level of the curvature dimension; instead of directly mapping the deceleration to the original curvature value. This suppresses jitter at the scene classification level, preventing jitter from being propagated down to the braking deceleration output.

[0074] In the above embodiments of this application, the look-ahead window adopts an adaptive arc length linked to time and vehicle speed, which is different from fixed-distance look-ahead; the window uses peak multi-point averaging to prioritize capturing the most dangerous curve conditions within the window, overcoming the defect that simple averaging will smooth out local sharp curves; an independent hysteresis state machine is configured for the curvature dimension to eliminate boundary jitter during the scene classification stage; the scene state is stabilized first, and then the deceleration level is mapped to prevent jitter from being transmitted to the braking output; it is specifically designed for fault-degraded parking scenarios, and when a fault occurs, the curve ahead is identified in advance, providing conservative constraints for subsequent braking control.

[0075] According to the embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scene dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road friction hysteresis mechanism further includes: determining multi-level lateral slope intervals using the vehicle roll angle as a lateral slope estimate; comparing the lateral slope interval hysteresis, gentle slope threshold, and steep slope threshold with the real-time vehicle roll angles in the flat road area, gentle slope area, and steep slope area of ​​the obtained multi-level lateral slope intervals to obtain lateral slope comparison results; completing the state transition between the flat road state, gentle slope state, and steep slope state in the multi-level lateral slope intervals based on the lateral slope comparison results to obtain the lateral slope interval with no jitter in the boundary switching area; and mapping the current state information of the lateral slope interval with no jitter in the boundary switching area to obtain the deceleration under the lateral slope dimension scene.

[0076] In one specific implementation, the deceleration under the cross-slope dimension scenario is obtained using a cross-slope interval hysteresis state machine:

[0077] Using the vehicle roll angle (i.e., the body roll angle) output by the positioning module as the cross slope estimate, a three-level cross slope range is defined:

[0078] Pinglu District: ;

[0079] Gentle slope area: ;

[0080] Steep slope area: .

[0081] The state transition rule also includes a hysteresis in the cross slope interval. As shown in formulas (7) and (8):

[0082] (7),

[0083] (8).

[0084] Cross-slope scenarios only occur when the vehicle speed exceeds a speed threshold. When the deceleration limit is activated, the lateral force caused by the cross slope is small at low speeds and does not pose a threat of rollover.

[0085] Cross slopes (lateral inclination of the road surface, often referred to as road superelevation in engineering) exert lateral forces on vehicles. At high speeds, cross slopes increase the risk of vehicle rollover and lateral slippage. Existing autonomous driving solutions for handling cross slopes have several problems.

[0086] Using the original cross slope estimate directly for threshold judgment lacks a hysteresis anti-shake mechanism: the vehicle roll angle output by the positioning module is affected by road bumps, sensor noise, and vibration interference. When the actual cross slope falls just near the grading threshold, the original measurement value will fluctuate slightly, causing the scene state to repeatedly switch between flat road, gentle slope, and steep slope; this further causes frequent jumps in the recommended deceleration level in the cross slope dimension, and repeated oscillations in braking output during fault parking, impairing parking stability.

[0087] Regardless of vehicle speed, the deceleration limitation due to the cross slope is applied throughout the entire process: the rollover moment caused by the cross slope is strongly correlated with vehicle speed: at high speeds, the inertia from the cross slope is superimposed, significantly increasing the risk of rollover; at low speeds, the vehicle's lateral inertia is very small, and the cross slope poses almost no risk of rollover. Many existing algorithms apply cross slope constraints regardless of vehicle speed: at low speeds, this unnecessarily excessively restricts braking intensity, resulting in an unreasonably longer stopping distance and slower response to malfunctions.

[0088] A single state machine is used for multiple scenarios: cross slope, curvature, and friction are coupled and processed in the same state machine. Noise in one dimension can interfere with the scenario judgment of all other dimensions, resulting in poor robustness. This invention configures the cross slope as a completely independent hysteresis state machine, decoupling it from the curvature and low-friction road surface state machines.

[0089] The scenario addressed in this application is that lateral control is locked when the vehicle is stopped due to a malfunction, and the steering wheel angle is no longer adjusted. If the braking deceleration is set too high under cross slope conditions, the lateral force generated by the cross slope will couple with the braking force, which will exacerbate the risk of vehicle instability and rollover. The reliability requirements for cross slope recognition-anti-shake-conditional activation are much higher than those for normal cruising.

[0090] This application does not necessarily apply deceleration limits simply because a steep slope is identified: only when the vehicle speed exceeds a preset high-speed threshold is the deceleration limit logic for the cross slope dimension activated, and the recommended deceleration level corresponding to the current cross slope scene state is mapped to participate in subsequent multi-dimensional scene optimization; when the vehicle speed is below the high-speed threshold, even if a steep slope condition is identified, the deceleration constraint corresponding to the cross slope is not activated.

[0091] It should be noted that at low speeds, the lateral inertial moment of the vehicle is very small, and the risk of rollover from the cross slope is negligible, so braking performance should not be sacrificed. Only at high speeds, when the lateral risk becomes significant, is a conservative deceleration constraint necessary. The stable cross slope scenario state output by the hysteresis state machine in the cross slope interval is used to obtain a recommended deceleration level independent of the cross slope dimension through a lookup table. This deceleration level is then fed into the multi-dimensional scenario fusion module, where it participates in the selection process along with the recommended deceleration levels for the curvature and road friction dimensions, ultimately choosing the most conservative value to obtain the expected scenario deceleration.

[0092] According to the embodiments of this application, the above-mentioned independent execution of deceleration acquisition operations under different scenario dimensions using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road surface friction hysteresis mechanism further includes: comparing the road surface friction hysteresis amount and the preset friction threshold with the current road surface friction coefficient to obtain the road surface friction comparison result; determining the current road surface state information based on the road surface friction comparison result, and using the current road surface state information to map the deceleration under the road surface friction dimension scenario.

[0093] In one specific implementation, the low-friction road surface hysteresis determination is performed by estimating the road surface friction coefficient output by the system identification module. Low attachment is determined by dual threshold hysteresis: Entering the low attachment state: Exiting the low-attached state: .

[0094] Furthermore, multi-dimensional scene fusion and priority arbitration: After independently evaluating the scenes in three dimensions—curvature, cross slope, and friction—this application classifies the scenes according to the following priority chain: Simultaneously, each dimension independently maps to its recommended deceleration level: the final expected deceleration is the one with the smallest absolute value among all recommended dimensions. This design ensures that the system can obtain a sufficiently conservative braking strategy when any single hazard dimension is activated.

[0095] According to an embodiment of this application, the above-mentioned friction circle constraint processing for vehicle longitudinal control using the road surface friction coefficient and the vehicle's current lateral acceleration to obtain the maximum longitudinal deceleration includes: determining the maximum longitudinal deceleration of the vehicle in the longitudinal direction using a friction circle dynamics model under the current road surface friction coefficient and the vehicle's current lateral acceleration; performing adhesion constraint processing on the desired deceleration of the scenario to calculate the safe limit of longitudinal deceleration allowed by the tire grip force.

[0096] In one specific implementation, the friction circle dynamics constraint is as follows: after determining the desired deceleration of the scene, the friction circle model is further used to constrain the maximum executable longitudinal deceleration, as shown in formula (9):

[0097] (9).

[0098] in: This is an estimated value for the current road surface friction coefficient. m / s² is the acceleration due to gravity. This is the current measured value of the vehicle's lateral acceleration.

[0099] The current lateral acceleration has already consumed part of the friction margin, so the limit that can be used for longitudinal braking is constrained by this.

[0100] Furthermore, the final arbitration of deceleration: the expected deceleration for the scenario. and friction circle limit Arbitration shall be conducted as shown in formula (10):

[0101] (10).

[0102] That is, the deceleration with the smaller absolute value is used to ensure that it does not exceed any constraint boundary.

[0103] Furthermore, horizontal control strategies

[0104] During a stop-and-go emergency, lateral control employs a strategy that maintains the current steering wheel angle.

[0105] The steering angle command at the moment the fault is triggered is locked, and the lateral control target is no longer updated.

[0106] Avoid superimposing steering actions during vehicle braking to prevent stability degradation caused by steering-braking coupling.

[0107] The vehicle decelerates along its current course until it comes to a stop due to inertia.

[0108] According to an embodiment of this application, the above-mentioned automatic vehicle parking method further includes: when the received parking downgrade instruction is a slow stop mode representing a non-emergency fault scenario, performing parking deceleration control on the vehicle according to a preset deceleration rate until the real-time vehicle speed meets a preset parking speed threshold, thereby completing the automatic parking of the vehicle in the slow stop mode.

[0109] In one specific implementation, the parking status determination and multi-level degradation mode are as follows: This application supports two fault parking modes: Emergency Stop Mode (EMERGENCY_STOP_HEADING): Executes the complete multi-dimensional arbitration logic described above to quickly stop the vehicle at the maximum safe deceleration. Slow Stop Mode (SLOW_STOP_HEADING): Decelerates slowly at a fixed GENTLE (gentle deceleration) level, suitable for non-emergency fault scenarios.

[0110] The determination of parking completion is shown in formula (11):

[0111] (11).

[0112] in This is the parking speed threshold.

[0113] The following describes another specific embodiment in conjunction with the appendix. Figure 3 This application provides a more detailed explanation of the automatic vehicle parking method.

[0114] Figure 3 The flowchart illustrates an arbitration scheme for automatic driving fault-safe parking deceleration based on multi-dimensional scene recognition and friction circle constraints according to an embodiment of this application.

[0115] like Figure 3 As shown, it continuously checks whether an FDIAG (Fault Diagnosis) stop command is received; when no FDIAG stop command is received, the vehicle operates in the normal control link and executes the normal vehicle control logic.

[0116] Upon receiving the FDIAG stop command, the system further determines the current degrading mode, which is divided into slow stop mode and emergency stop mode.

[0117] If the system is determined to be in slow stop mode, the system will output GENTLE deceleration while maintaining steering control, and then proceed to the vehicle speed determination stage.

[0118] If the emergency stop mode is determined, the vehicle status, path information, positioning results, and identification results are read, and the working condition processing of curvature dimension, cross slope dimension, and road friction dimension is carried out in parallel.

[0119] In the curvature dimension processing flow: calculate the look-ahead arc length window, collect the curvature data inside the look-ahead arc length window and perform sorting processing; select the top multiple groups of peak curvature after sorting to calculate the mean, convert the mean result to obtain the equivalent radius, and use the equivalent radius to update the curvature hysteresis state machine.

[0120] In the cross slope dimension processing flow: read the vehicle roll angle data and update the cross slope hysteresis state machine using the vehicle roll angle.

[0121] In the road surface friction dimension processing flow: read the friction coefficient estimation results and update the low friction hysteresis state.

[0122] After completing the state machine update for each of the three dimensions, the deceleration level of the corresponding dimension is obtained by independently mapping the scene classification state output of each dimension; the deceleration with the smallest absolute value is selected as the expected deceleration of the scene.

[0123] Perform longitudinal limit calculation of the friction circle to obtain the safe limit of longitudinal deceleration allowed by tire grip; perform optimal arbitration between the expected deceleration of the scenario and the longitudinal limit of the friction circle, select the deceleration with the smaller absolute value, output the deceleration and maintain the steering angle.

[0124] After outputting the data, determine whether the vehicle's real-time speed is less than the parking threshold.

[0125] If the vehicle's real-time speed does not meet the condition of being less than the stopping threshold, the next control cycle will continue, and the vehicle status, path, location, and identification results will be read again, and all the emergency stop mode processing logic will be executed in a loop.

[0126] If the vehicle's real-time speed is less than the parking threshold, a parking completion flag will be output.

[0127] After outputting GENTLE deceleration and maintaining steering in slow stop mode, the vehicle speed judgment logic is entered again; if the vehicle speed meets the parking threshold condition, a parking completion flag is output; otherwise, the loop continues.

[0128] The automatic vehicle parking method provided in this application achieves multi-dimensional scene perception fusion: unlike a single deceleration strategy, this application simultaneously considers three dimensions: road curvature, cross slope angle, and road surface friction coefficient. Each dimension is independently evaluated and the most conservative value is taken to ensure safety under any single dangerous scenario. It enables forward-looking peak curvature analysis: instead of using the instantaneous curvature at the current position, it is based on the peak curvature statistics within a preview window of 1-8 seconds ahead, providing early perception of curves ahead and solving the problem of scene judgment lag. It features a full-dimensional hysteresis state machine: independent hysteresis mechanisms are used for state updates in curvature intervals, cross slope intervals, and low-friction states, eliminating deceleration target oscillations caused by boundary switching and improving braking smoothness. It has a friction circle dynamics constraint fallback: physical limit constraints based on the vehicle dynamics friction circle model are superimposed on the scene-level deceleration decision to ensure that the braking force does not exceed the tire adhesion limit under any operating condition. It supports multi-level degradation modes: emergency stop and slow stop modes are distinguished according to the severity of the fault, providing flexible degradation means for the system safety architecture.

[0129] Based on the above-described automatic vehicle parking method, embodiments of this application also provide an automatic vehicle parking device. The following will be combined with... Figure 4 The device is described in detail.

[0130] Figure 4 A schematic block diagram of a vehicle automatic parking device according to an embodiment of this application is shown.

[0131] like Figure 4 As shown, the vehicle automatic parking device 400 in this embodiment includes a multi-dimensional scene perception module 410, a friction circle constraint module 420, a deceleration arbitration module 430, and a parking deceleration module 440.

[0132] The multi-dimensional scene perception module 410 is used to independently perform deceleration acquisition operations under different scene dimensions when the received parking downgrade command is an emergency stop mode, utilizing the look-ahead path curvature interval hysteresis mechanism, the vehicle body roll angle lateral slope interval hysteresis mechanism, and the road surface friction hysteresis mechanism, and to perform a deceleration arbitration operation on the deceleration under different scene dimensions to obtain the desired deceleration for the scene. In one embodiment, the multi-dimensional scene perception module 410 can be used to execute step S210 described above, which will not be repeated here.

[0133] The friction circle constraint module 420 is used to perform friction circle constraint processing on the longitudinal control of the vehicle using the road surface friction coefficient and the vehicle's current lateral acceleration to obtain the maximum longitudinal deceleration. In one embodiment, the friction circle constraint module 420 can be used to execute step S220 described above, which will not be repeated here.

[0134] The deceleration arbitration module 430 is used to perform a deceleration arbitration operation again between the desired deceleration and the maximum longitudinal deceleration to obtain the target braking command. In one embodiment, the deceleration arbitration module 430 can be used to execute step S230 described above, which will not be repeated here.

[0135] The parking deceleration module 440 is used to perform lateral control on the vehicle by locking the steering angle command at the moment of vehicle malfunction, and to perform parking deceleration control on the vehicle using the target braking command until the real-time vehicle speed meets the preset parking speed threshold, thus completing the automatic parking of the vehicle in emergency stop mode. In one embodiment, the parking deceleration module 440 can be used to execute step S240 described above, which will not be repeated here.

[0136] According to embodiments of this application, any multiple modules among the multi-dimensional scene perception module 410, friction circle constraint module 420, deceleration arbitration module 430, and parking deceleration module 440 can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the multi-dimensional scene perception module 410, friction circle constraint module 420, deceleration arbitration module 430, and parking deceleration module 440 can be at least partially implemented as hardware circuits, such as field-programmable gate arrays, programmable logic arrays, systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits, or any other reasonable means of integrating or packaging circuits, or implemented in hardware or firmware, or in any one of software, hardware, and firmware implementations, or in a suitable combination of any of these. Alternatively, at least one of the multi-dimensional scene perception module 410, friction circle constraint module 420, deceleration arbitration module 430, and parking deceleration module 440 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0137] Figure 5 A block diagram schematically illustrates an electronic device suitable for implementing an automatic vehicle parking method according to an embodiment of this application.

[0138] like Figure 5As shown, an electronic device 500 according to an embodiment of this application includes a processor 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory 502 or a program loaded from a storage portion 508 into a random access memory 503. The processor 501 may include, for example, a general-purpose microprocessor, an instruction set processor and / or an associated chipset and / or a dedicated microprocessor. The processor 501 may also include onboard memory for caching purposes. The processor 501 may include a single processing unit or multiple processing units for executing different steps of the method flow according to an embodiment of this application.

[0139] Random access memory 503 stores various programs and data required for the operation of electronic device 500. Processor 501, read-only memory 502, and random access memory 503 are interconnected via bus 504. Processor 501 executes various steps of the method flow according to embodiments of this application by executing programs in read-only memory 502 and / or random access memory 503. It should be noted that the programs may also be stored in one or more memories other than read-only memory 502 and random access memory 503. Processor 501 may also execute various steps of the method flow according to embodiments of this application by executing programs stored in said one or more memories.

[0140] According to embodiments of this application, the electronic device 500 may further include an input / output interface 505, which is also connected to a bus 504. The electronic device 500 may also include one or more of the following components connected to the input / output interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube, liquid crystal display, etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card, such as a local area network card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the input / output interface 505 as needed. A removable medium 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 510 as needed so that computer programs read from it can be installed into the storage section 508 as needed.

[0141] Embodiments of this application also provide a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0142] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In embodiments of this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include the read-only memory 502, and / or random access memory 503, and / or one or more memories other than read-only memory 502 and random access memory 503 described above.

[0143] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to cause the computer system to implement the methods provided in the embodiments of this application.

[0144] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and may be downloaded and installed via the communication section 509, and / or installed from a removable medium 511. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0145] In embodiments of this application, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by processor 501, it performs the functions defined in the system of this application embodiment. According to embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0146] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0147] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0148] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for automatically parking a vehicle, characterized in that, The method includes: When the received parking downgrade command is an emergency stop mode, the deceleration acquisition operation under different scenario dimensions is independently executed by using the forward path curvature interval hysteresis mechanism, the vehicle body roll angle lateral slope interval hysteresis mechanism and the road surface friction hysteresis mechanism, and the deceleration under different scenario dimensions is arbitrated to obtain the scenario expected deceleration. By using the road surface friction coefficient and the vehicle's current lateral acceleration to apply friction circle constraints to the vehicle's longitudinal control, the maximum longitudinal deceleration is obtained. The desired deceleration in the scenario is again compared with the maximum longitudinal deceleration to obtain the target braking command. The vehicle is laterally controlled by locking the steering angle command at the moment of the vehicle malfunction, and the vehicle is stopped and decelerated by using the target braking command until the real-time speed of the vehicle meets the preset stopping speed threshold, thus completing the automatic stopping of the vehicle in the emergency stop mode.

2. The method according to claim 1, characterized in that, The deceleration acquisition operations under different scenario dimensions are performed independently using the look-ahead path curvature interval hysteresis mechanism, the vehicle roll angle lateral slope interval hysteresis mechanism, and the road friction hysteresis mechanism, including: The forward arc length window is determined based on the vehicle's current speed, and curvature data falling within the forward arc length window is extracted from the vehicle's planned path. The extracted curvature data is sorted and averaged to obtain the mean curvature value, and the mean curvature value is used to determine the look-ahead equivalent radius. The forward equivalent radius is mapped to a multi-level curvature interval, and the state transition of the multi-level curvature interval is controlled by the curvature interval hysteresis to obtain a multi-level curvature interval without jitter in the boundary switching region. The deceleration in the curvature dimension scenario of the look-forward path is obtained by mapping the current state information of the multi-level curvature interval under the condition of no jitter in the boundary switching region.

3. The method according to claim 2, characterized in that, The extracted curvature data is sorted and averaged to obtain the mean curvature values, including: Perform a descending sort operation on the absolute values ​​of the extracted curvature data to obtain the descending sort result; From the descending sort results, select the top-ranked absolute curvature values ​​and perform an averaging operation on the multiple absolute curvature values ​​to obtain the mean curvature value.

4. The method according to claim 2, characterized in that, By controlling the state transition of the multi-level curvature interval using curvature interval hysteresis, the following multi-level curvature intervals with no jitter in the boundary switching region are obtained: The curvature interval hysteresis, the curvature threshold of gentle curves and the curvature threshold of sharp curves are compared with the curvature of straight road area, curvature of gentle curve area and curvature of sharp curve area in the multi-level curvature interval respectively to obtain curvature comparison results. Based on the curvature comparison results, the state transitions between straight road, gentle curve, and sharp curve states in the multi-level curvature interval are completed, resulting in a multi-level curvature interval without jitter in the boundary switching area.

5. The method according to claim 2, characterized in that, Also includes: The vehicle roll angle, used as a cross slope estimate, is used to determine the multi-level cross slope range; The hysteresis, gentle slope threshold, and steep slope threshold of the cross slope interval are compared with the vehicle roll angle in the flat area, the vehicle roll angle in the gentle slope area, and the real-time vehicle roll angle in the steep slope area of ​​the multi-level cross slope interval to obtain the cross slope comparison results. Based on the cross slope comparison results, the state transition between flat road state, gentle slope state and steep slope state in the multi-level cross slope interval is completed to obtain the cross slope interval with no shaking in the boundary switching area. The deceleration in the cross-slope dimension scenario is obtained by mapping the current state information of the cross-slope interval in the boundary switching region without jitter.

6. The method according to claim 2, characterized in that, Also includes: The road surface friction hysteresis and preset friction threshold are compared with the current road surface friction coefficient to obtain the road surface friction comparison result; The current road surface state information is determined based on the road surface friction comparison results, and the deceleration under the road surface friction dimension is obtained by mapping the current road surface state information.

7. The method according to claim 1, characterized in that, By applying friction circle constraints to the vehicle's longitudinal control using the road surface friction coefficient and the vehicle's current lateral acceleration, the maximum longitudinal deceleration is obtained, including: Given the current road surface friction coefficient and the current lateral acceleration of the vehicle, the maximum longitudinal deceleration of the vehicle in the longitudinal direction is determined using the friction circle dynamics model. By applying adhesion constraints to the desired deceleration in the scenario, the safe limit of longitudinal deceleration allowed by tire grip is calculated.

8. The method according to any one of claims 1 to 7, characterized in that, Also includes: When the received parking downgrade command is a slow stop mode representing a non-emergency fault scenario, the vehicle is controlled to stop and decelerate according to a preset deceleration until the real-time speed of the vehicle meets the preset parking speed threshold, thus completing the automatic parking of the vehicle in the slow stop mode.

9. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 7.

Images