Parking control method and device and unmanned vehicle

Abstract

The invention provides a parking control method and device and an unmanned vehicle. The method comprises the steps that driving information of the unmanned vehicle driving to a preset area range of a parking point is acquired; determining a current parking control stage of the unmanned vehicle according to the driving information; according to the parking control strategy corresponding to the parking control stage of the unmanned vehicle, the driving state of the unmanned vehicle is controlled, and parking is completed.

Description

Technical Field

[0001] This disclosure relates to the fields of autonomous driving and vehicle control technology, and in particular to a parking control method, device and unmanned vehicle. Background Technology

[0002] With the iterative upgrades of autonomous driving technology, unmanned vehicles have become core transportation equipment for improving operational efficiency and reducing labor costs. In various operational processes, precise docking is a crucial step for unmanned vehicles, determining operational quality, process continuity, and operational safety. Therefore, to adapt to the core needs of material loading and unloading, personnel transfer, and task handover in different scenarios, stringent requirements have been placed on docking accuracy, driving stability, and safety redundancy.

[0003] However, various types of unmanned vehicles generally face challenges in precise parking control due to their inherent characteristics. This is especially true for medium and large-sized unmanned vehicles, which are large in size and heavy in weight, with significant motion inertia. Furthermore, the speed control actuators have unavoidable response delays. The combination of these factors significantly increases the difficulty of controlling the vehicle's driving status, making it prone to problems such as speed fluctuations, position deviations, overshooting, or failure to reach the parking point, thus failing to meet the requirements for high-precision parking. Summary of the Invention

[0004] This disclosure provides a parking control method, device, and unmanned vehicle to solve the problem that existing unmanned vehicles cannot accurately park at designated stops.

[0005] In view of the above problems, firstly, the present disclosure provides a parking control method, including: Obtain driving information of the unmanned vehicle when it reaches the preset area of ​​the parking point; Based on the driving information, the current parking control stage of the unmanned vehicle is determined; The driving state of the unmanned vehicle is controlled according to the parking control strategy corresponding to the parking control stage in which the unmanned vehicle is located, so as to complete the parking.

[0006] In conjunction with the first aspect, in one possible implementation, the driving information includes the trajectory length between the unmanned vehicle and the stop point; Based on the driving information, the current parking control phase of the autonomous vehicle is determined, including: The parking control zone where the unmanned vehicle is currently located is determined based on the trajectory length between the unmanned vehicle and the parking point; Based on the parking control zone and the parking control strategy corresponding to the parking control zone, the speed control mode of the driverless car is determined. The step of controlling the driving state of the autonomous vehicle according to the parking control strategy corresponding to the parking control stage in which the autonomous vehicle is located includes: The driving state of the unmanned vehicle is controlled according to the vehicle speed control mode.

[0007] In conjunction with the first aspect, in one possible implementation, the parking control zone is a continuous distance interval that pre-divides the preset area range; the parking control zone includes: a first sub-region range, a second sub-region range, and a third sub-region range on the decision trajectory of the unmanned vehicle, starting from the parking point and extending from the parking point to the current position of the unmanned vehicle.

[0008] In conjunction with the first aspect, in one possible implementation, the vehicle speed control mode includes: a speed control mode following the decision trajectory point, a position following mode, and an open-loop deceleration mode. The step of determining the vehicle speed control mode of the autonomous vehicle based on the parking control zone and the corresponding parking control strategy includes: Based on the driving information of the unmanned vehicle, determine the deviation between the expected driving state and the current driving state of the unmanned vehicle; When the unmanned vehicle is within the third sub-region, it is determined that the unmanned vehicle adopts the following decision trajectory point speed control mode; the following decision trajectory point speed control mode is used to control the speed of the unmanned vehicle based on the expected speed of the unmanned vehicle at the trajectory point. During the autonomous vehicle's operation within the second sub-region, if the deviation value is greater than or equal to a second preset value, the autonomous vehicle is determined to adopt a following decision trajectory point speed control mode; if the deviation value is less than the second preset value, the autonomous vehicle is determined to adopt a position following mode; the position following mode is used to control the speed of the autonomous vehicle based on the trajectory length between the autonomous vehicle and the stop point and a preset minimum controllable speed. During the autonomous vehicle's operation within the first sub-region, if the deviation value is greater than or equal to a first preset value, the autonomous vehicle is determined to adopt a speed control mode that follows the decision trajectory point; if the deviation value is less than the first preset value, the autonomous vehicle is determined to adopt an open-loop deceleration mode; the open-loop deceleration mode is used to control the deceleration of the autonomous vehicle based on real-time vehicle speed, distance tolerance, and road slope.

[0009] In conjunction with the first aspect, in one possible implementation, determining the deviation between the desired driving state and the current driving state of the autonomous vehicle based on its driving information includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the unmanned vehicle is determined based on the real-time vehicle speed, the desired acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance; The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. The symbolic features of the sliding surface are extracted using a symbolic function to determine the deviation between the desired driving state and the current driving state of the unmanned vehicle.

[0010] In conjunction with the first aspect, in one possible implementation, when the autonomous vehicle is within the third sub-region and during the autonomous vehicle's operation within the second sub-region, after determining that the autonomous vehicle adopts a following decision trajectory point speed control mode, controlling the autonomous vehicle's driving state according to the speed control mode includes: The vehicle speed control mode following the decision trajectory point is adopted to control the unmanned vehicle to follow the decision trajectory point at a first speed; the first speed is not less than the preset minimum controllable speed. During the autonomous vehicle's operation within the first sub-region, after determining the following decision trajectory point speed control mode adopted by the autonomous vehicle, controlling the autonomous vehicle's driving state according to the speed control mode includes: Based on the driving information of the unmanned vehicle, the first target acceleration of the unmanned vehicle is determined; wherein, the driving information includes: the trajectory length between the unmanned vehicle and the stopping point, the real-time vehicle speed, and the road gradient; Based on the first target acceleration, the unmanned vehicle is controlled to decelerate and stop.

[0011] In conjunction with the first aspect, in one possible implementation, after determining that the unmanned vehicle is using an open-loop deceleration mode while it is driving within the first sub-region, controlling the driving state of the unmanned vehicle according to the speed control mode includes: Based on the driving information of the unmanned vehicle, the second target acceleration of the unmanned vehicle is determined, including: The vehicle speed component is determined by the vehicle speed coefficient and the square of the real-time vehicle speed. The distance component is determined based on the reciprocal of the sum of the trajectory length and the distance tolerance, and the distance coefficient; The slope components are determined based on the sine of the road slope, gravitational acceleration, and slope coefficient. The second target acceleration of the unmanned vehicle is determined based on the vehicle speed component, the distance component, and the slope component. Based on the second target acceleration, the unmanned vehicle is controlled to decelerate and stop.

[0012] In conjunction with the first aspect, in one possible implementation, after determining that the unmanned vehicle is in position-following mode while traveling within the second sub-region, controlling the driving state of the unmanned vehicle according to the speed control mode includes: Based on the deviation value, determine the acceleration feedforward of the unmanned vehicle; The acceleration feedforward and the desired acceleration of the unmanned vehicle are superimposed to control the unmanned vehicle to travel at a second speed that is not less than a preset minimum controllable speed, and the difference between the second speed and the preset minimum controllable speed is within a first preset threshold range.

[0013] In conjunction with the first aspect, in one possible implementation, determining the acceleration feedforward of the autonomous vehicle based on the deviation value includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the unmanned vehicle is determined based on the real-time vehicle speed, the desired acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance; The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. The sign features of the sliding surface are extracted by the sign function to determine the deviation between the expected driving state and the current driving state of the unmanned vehicle. The sliding surface adaptive coefficient is determined based on the deviation value, the preset adaptive coefficient, and the vehicle speed control execution cycle. The acceleration feedforward of the unmanned vehicle is determined based on the deviation value and the sliding surface adaptive coefficient.

[0014] In conjunction with the first aspect, in one possible implementation, the method further includes: when the trajectory length information indicates a change in the location of the stop point, switching the vehicle speed control mode according to the driving information, and controlling the driving state of the unmanned vehicle according to the switched vehicle speed control mode; The step of switching the vehicle speed control mode based on the driving information includes one or a combination of the following: If the current vehicle speed control mode is the location following mode, and the driving information meets the first condition, the current vehicle speed control mode is switched to the following decision trajectory point speed control mode; the first condition includes: the trajectory length is greater than a first preset distance threshold; or, the trajectory length is between a second preset distance threshold and a first preset distance threshold, and the duration exceeds a preset time threshold; or, the real-time vehicle speed is less than or equal to a first vehicle speed threshold; the first preset distance threshold is greater than the second preset distance threshold; If the current vehicle speed control mode is position following mode, and the driving information meets the second condition, the current vehicle speed control mode is switched to open-loop deceleration mode; the second condition includes: the trajectory length is less than the third preset distance threshold, and the deviation value is less than the first preset value; or, the trajectory length is less than the fourth preset distance threshold; the third preset distance threshold is greater than the fourth preset distance threshold. If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the third condition, the current vehicle speed control mode is switched to the position following mode; the third condition includes: the trajectory length is between the third preset distance threshold and the second preset distance threshold, the real-time vehicle speed is greater than the first vehicle speed threshold, the unmanned vehicle is not in the starting state, the deviation value is less than the second preset value, and the current expected acceleration is less than the preset acceleration threshold. If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the fourth condition, the current vehicle speed control mode is switched to the open-loop deceleration mode; the fourth condition includes: the trajectory length is less than the third preset distance threshold, the unmanned vehicle is not in a starting state, the deviation value is less than the first preset value, and the real-time vehicle speed is between the first vehicle speed threshold and the second vehicle speed threshold; or, the trajectory length is less than the fourth preset distance threshold, and the real-time vehicle speed is greater than the first vehicle speed threshold. If the current vehicle speed control mode is open-loop deceleration mode, and the driving information meets the fifth condition, the current vehicle speed control mode will be switched to the following decision trajectory point speed control mode; the fifth condition includes: the real-time vehicle speed is less than or equal to the first vehicle speed threshold, or the trajectory length is greater than the second preset distance threshold.

[0015] In conjunction with the first aspect, in one possible implementation, the driving information includes the trajectory length between the unmanned vehicle and the stop point; The acquisition of driving information of the unmanned vehicle within the preset area of ​​the parking point includes: The distance between the unmanned vehicle and the parking point is obtained using a preset device, and this distance is used as the trajectory length between the unmanned vehicle and the parking point; the preset device includes: a sensing device and / or a positioning device; If the absolute value of the difference between the current trajectory length and the trajectory length at the previous moment is greater than the second preset threshold, the remaining length of the decision trajectory point will be used as the new trajectory length.

[0016] In a second aspect, a parking control device is provided, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the parking control device is in operation, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the steps of the parking control method as described in the first aspect or in any possible embodiment of the first aspect are performed.

[0017] Thirdly, an unmanned vehicle is provided, including: a parking control device as described in the second aspect.

[0018] The beneficial effects of the embodiments disclosed herein include: This disclosure provides a parking control method, device, and unmanned vehicle, comprising: acquiring driving information of the unmanned vehicle within a preset area of ​​a parking point; determining the current parking control stage of the unmanned vehicle based on the driving information; and controlling the driving state of the unmanned vehicle according to the parking control strategy corresponding to the current parking control stage to complete parking. The parking control method provided in this disclosure is based on a phased control strategy and is specifically adapted to the inherent characteristics of unmanned vehicles, such as high inertia and large delay of speed control actuators. By setting up parking control phases and corresponding parking control strategies, the random differences in initial vehicle speed and deceleration before parking are transformed into relatively fixed vehicle states, and the vehicle posture is precisely calibrated simultaneously, reducing the difficulty of subsequent parking control from the source. The customized deceleration parking control method replaces the traditional speed following control method, which not only effectively solves the technical problem of insufficient speed control accuracy of unmanned vehicles in the low-speed range, but also stably achieves high-precision parking, meeting the stringent process requirements of unloading scenarios such as mine crushing stations and spoil heaps, and fundamentally eliminating the phenomenon of collision with retaining walls; it can also effectively reduce vehicle speed fluctuations and vehicle impacts during parking, reduce equipment wear, avoid secondary start-up adjustments, and significantly improve the safety and overall stability of mining operations. Attached Figure Description

[0019] Figure 1 A flowchart of a parking control method provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram of parking control zones provided in an embodiment of the present disclosure; Figure 3 This is a schematic diagram illustrating the switching of vehicle speed control modes provided in an embodiment of this disclosure. Detailed Implementation

[0020] This disclosure provides a parking control method, apparatus, and unmanned vehicle. Preferred embodiments of this disclosure are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustrative and explanatory purposes only and are not intended to limit the scope of this disclosure. Furthermore, the embodiments and features described herein can be combined with each other unless otherwise specified.

[0021] This disclosure provides a parking control method, such as... Figure 1 As shown, it includes: S101. Obtain driving information of the unmanned vehicle within the preset area of ​​the parking point; S102. Based on the driving information, determine the current parking control stage of the unmanned vehicle; S103. Based on the parking control strategy corresponding to the parking control stage in which the driverless vehicle is located, control the driving state of the driverless vehicle to complete the parking process.

[0022] This disclosure applies to various scenarios such as smart mines, industrial park transfers, port loading and unloading, logistics warehousing, and municipal operations. In mining and material transfer production scenarios, unmanned vehicles are typically driverless mining trucks. As core transportation equipment, driverless mining trucks undertake the critical task of material transfer. Due to the special nature of the operating scenarios, driverless mining trucks are designed with a high self-weight and large body size. Their empty weight can reach over 40 tons, and their weight under full load can reach over 140 tons. The body length can reach over 10 meters, and the width can reach over 5 meters. This structural feature gives driverless mining trucks a great deal of motion inertia, and the vehicle speed control actuator has a significant response delay. These inherent characteristics become the core technical obstacle to achieving precise parking of driverless mining trucks in unloading scenarios, and precise parking in this scenario has always been a technical problem that urgently needs to be solved in the industry.

[0023] Taking the unloading scenarios of crushing stations in sand and gravel aggregate mines and coal mines as typical examples, mine operators have put forward stringent process requirements for the parking accuracy of unmanned mining trucks. Typically, the distance tolerance between the unmanned mining truck and the parking point is required to be controlled within 15 centimeters. In some high-precision scenarios, the distance tolerance is even limited to within 10 centimeters. In addition, to ensure equipment safety and operational order, it is strictly forbidden for unmanned mining trucks to collide with retaining walls.

[0024] Traditionally, speed-following control is commonly used. This method uses a preset desired speed as the control benchmark. It collects real-time vehicle speed data, calculates the deviation between the real-time speed and the desired speed, and dynamically adjusts the vehicle's driving state based on closed-loop control logic. This ensures that the real-time speed consistently and smoothly converges to the desired speed, making it the mainstream speed control method for current autonomous vehicles. However, this method only focuses on the speed-following effect. While it can meet basic control requirements in scenarios such as conventional road driving and long-distance mining transport, it has significant adaptability limitations when applied to high-precision parking scenarios such as unloading unmanned mining trucks. When an unmanned mining truck approaches a stopping point, the expected speed is usually extremely low or even zero. At this time, closed-loop control based on the expected speed is prone to problems such as low signal-to-noise ratio and amplified response delay of the speed control actuator, resulting in insufficient control accuracy and unstable vehicle driving status. For example, in parking scenarios at mine stopping points such as crushing stations and spoil heaps, speed-following closed-loop control is very likely to cause unmanned mining trucks to crash into retaining walls and have excessive parking position deviations. In some cases, the vehicle needs to start again to adjust the parking position, which not only seriously reduces the production efficiency of mine material transfer, but also poses a great threat to the safety of operating equipment. Moreover, this method cannot be directly adapted to the high positional accuracy stopping requirements in mining scenarios and is difficult to meet the process requirements of mine production.

[0025] In this embodiment, by collecting driving information of the unmanned vehicle entering the preset area of ​​the parking point, dividing the parking control stage based on the driving information, and finally matching the exclusive parking control strategy for each stage, the vehicle speed and position are controlled in stages and differentiatedly. This effectively adapts to the inherent characteristics of the unmanned vehicle's high inertia and large actuator delay, solves the problem of high-precision parking in the unloading scenario, and ensures the accuracy, safety and stability of the parking process.

[0026] When the unmanned vehicle (UAV) reaches a preset area (typically 5-20 meters, dynamically adjustable based on UAV performance and mine conditions) from its stopping point (such as the crushing station retaining wall or unloading area), a data acquisition mechanism is triggered. This mechanism collects multi-dimensional driving information of the UAV at high frequency and with high precision, providing accurate data support for subsequent parking control phase determination and parking control strategy matching. Driving information, as the perception foundation of the entire parking control process, can include multi-dimensional data, such as: real-time vehicle speed, current acceleration / deceleration, vehicle attitude (e.g., pitch angle, adapted for slope compensation), speed control actuator response status; current precise positioning of the UAV, trajectory length between the UAV and the stopping point, coordinates of the pre-aimed trajectory point and corresponding expected speed, road slope, road resistance coefficient, speed control execution delay time, sensor operating status, stopping point location coordinates, preset minimum controllable speed, distance tolerance, etc. Based on the UAV's driving information and preset phase division rules, the current parking control phase of the UAV is determined, achieving refined, layered control of the parking process. For example, in the first phase, long-distance speed following is performed. In the first stage, the autonomous vehicle's trajectory length to the parking point is greater than 4 meters. The core objective in this stage is to smoothly decelerate while maintaining trajectory following. In the second stage, close-range speed control is implemented. The trajectory length is between 0.8 meters and 4 meters, with the real-time speed approaching the preset minimum controllable speed. The core objective in this stage is to stabilize the speed within a controllable range, eliminating the randomness of the initial state before each stop. In the third stage, precise parking is achieved. The trajectory length is ≤0.8 meters. The core objective in this stage is precise positioning, meeting a distance tolerance requirement of 10-15cm, and strictly prohibiting collisions with barriers.

[0027] Furthermore, corresponding parking control strategies are set for each parking control stage. Based on driving information, a target acceleration is output to dynamically adjust the autonomous vehicle's driving state until it accurately stops at the target position. For example, in the first stage, a "following decision trajectory point speed control mode" is adopted. The desired speed of the pre-aimed trajectory point is used as the control target. The autonomous vehicle's speed is adjusted in real time through closed-loop control, while incorporating execution delay compensation and inertial buffer design to ensure that the autonomous vehicle decelerates smoothly along the planned trajectory, avoiding sudden speed drops that could cause vehicle impact. In the second stage, a "position following control mode" is adopted. The trajectory length between the autonomous vehicle and the stopping point is used as the core control basis. Combined with the weighted calculation results of position deviation and speed deviation, the vehicle's speed and body posture are finely adjusted to stabilize the real-time speed within the first preset threshold range of the preset minimum controllable speed. This eliminates random differences in initial speed and deceleration before each stop, converting the autonomous vehicle's state to a fixed initial state. In the third stage, an "open-loop deceleration control mode" is adopted to decelerate the autonomous vehicle, precisely control the parking distance, and ensure that the parking accuracy is within the distance tolerance range, without secondary start-up adjustments.

[0028] In this embodiment, the parking control method is designed separately from the general driving control method. Based on a phased management strategy, it is specifically adapted to the inherent characteristics of unmanned vehicles, such as high inertia and large delay of speed control actuators. The random differences in initial vehicle speed and deceleration before parking are transformed into a relatively fixed vehicle state, and the vehicle attitude is simultaneously and accurately calibrated, reducing the difficulty of subsequent parking control from the source. According to a dedicated deceleration and parking design, it replaces the traditional speed-following control method, effectively solving the problem of insufficient low-speed control accuracy of unmanned vehicles, stably achieving high-precision parking, meeting the stringent process requirements of unloading scenarios such as mine crushing stations and spoil heaps, and eliminating collisions with retaining walls. It also reduces speed fluctuations and vehicle impact during parking, reduces equipment wear, avoids secondary start-up adjustments, and significantly improves the safety and stability of mining operations.

[0029] In another embodiment of this disclosure, the driving information includes the trajectory length between the unmanned vehicle and the stop point; In step S102 above, the parking control stage of the autonomous vehicle is determined based on the driving information, including: Step 1: Determine the parking control zone where the unmanned vehicle is currently located based on the trajectory length between the unmanned vehicle and the parking point; Step 2: Determine the vehicle speed control mode based on the parking control zones and the corresponding parking control strategies for each zone; In step S103 above, the driving state of the autonomous vehicle is controlled according to the parking control strategy corresponding to the parking control stage in which the autonomous vehicle is located, including: Step 3: Control the driving status of the driverless car according to the vehicle speed control mode.

[0030] In this embodiment, driving information, including the trajectory length between the autonomous vehicle and the parking point, is collected. Based on the trajectory length, the parking control zone where the autonomous vehicle is located is determined, and a corresponding parking control strategy and speed control mode are matched. Finally, the driving state of the autonomous vehicle is controlled according to the speed control mode. The trajectory length between the autonomous vehicle and the parking point is the remaining driving path distance from the current position of the autonomous vehicle to the parking point. For step 1 above, multiple independent parking control zones and corresponding trajectory length thresholds are pre-set. The real-time trajectory length between the autonomous vehicle and the parking point is collected and compared with the trajectory length range of the preset parking control zones to determine the parking control zone to which the current position of the autonomous vehicle belongs, achieving layered control of the parking process. Each preset zone corresponds to a clear trajectory length range and a specific parking control target. Through accurate matching of the real-time trajectory length with the preset range, the current parking control zone is quickly locked, ensuring clear control priorities at different parking stages and laying the foundation for subsequent accurate matching of control strategies. For step 2 above, a corresponding parking control strategy is configured for each preset parking control zone. Based on the current parking control zone, the corresponding parking control strategy is invoked, and a speed control mode suitable for that zone and strategy is matched. Different parking control zones correspond to different parking control objectives, and the corresponding parking control strategies are designed around these objectives. The vehicle speed control mode serves as the execution vehicle for these strategies, ensuring that they are translated into specific speed control actions to adapt to different parking needs. For step 3 above, based on the matched speed control mode and the collected real-time driving information of the autonomous vehicle, corresponding speed control commands are generated and sent to the autonomous vehicle's speed control actuator to adjust its speed, acceleration, and other driving parameters, thus controlling its driving state. Using the trajectory length between the autonomous vehicle and the parking point as the core, precise division of parking control zones and scientific matching of speed control modes are achieved. The complex parking control process is broken down into multiple simple and controllable sub-stages, each corresponding to a clear control objective. By adapting specific control methods to different stages, the autonomous vehicle can achieve optimal control at different parking stages, effectively improving parking position accuracy and driving stability, ensuring a smooth and stable parking process.

[0031] In another embodiment of this disclosure, the parking control zone may refer to a continuous distance interval that is pre-divided into a preset area range; the parking control zone includes: a first sub-area range, a second sub-area range, and a third sub-area range on the decision trajectory of the unmanned vehicle, starting from the parking point and extending from the parking point to the current position of the unmanned vehicle.

[0032] In this embodiment of the disclosure, starting from the parking point, three consecutive zones—a first sub-region, a second sub-region, and a third sub-region—are clearly defined along the autonomous vehicle's decision trajectory toward its current position; these are the parking control zones. For example... Figure 2As shown, the horizontal axis represents the trajectory length between the autonomous vehicle and the parking point. Here, d1 is greater than d2, d2 is greater than d3, and d3 is greater than d4. d4 is the parking point location. Starting from the parking point, along the autonomous vehicle's planned decision trajectory towards its current location (i.e., the reverse extension of the parking path), three sub-regions are sequentially divided, with these three sub-regions being continuously connected. The first sub-region lies between d3 and d4, the second between d2 and d3, and the third between d1 and d2. The vertical axis represents the autonomous vehicle's speed, where solid lines represent the desired speed and dashed lines represent the real-time speed. Vmin represents the preset minimum controllable speed. As the autonomous vehicle approaches the parking point, it sequentially transitions from the third sub-region to the second sub-region, and finally to the first sub-region, achieving dynamic switching and precise positioning of the sub-regions. Each sub-region corresponds to a clearly defined distance range, thus breaking down the autonomous vehicle parking path into gradient and quantifiable control units, avoiding the ambiguity of unzoned control. This ensures that autonomous vehicles can obtain accurately adapted control logic in different locations, significantly improving the precision and standardization of parking control.

[0033] In another embodiment of this disclosure, the vehicle speed control mode includes: a vehicle speed control mode following the decision trajectory point, a position following mode, and an open-loop deceleration mode. In step 2 above, based on the parking control zones and the corresponding parking control strategies for each zone, the speed control mode of the autonomous vehicle is determined, including: Step 1: Based on the autonomous vehicle's driving information, determine the deviation between the expected driving state and the current driving state of the autonomous vehicle; Step 2: When the autonomous vehicle is within the third sub-region, determine that the autonomous vehicle adopts the following decision trajectory point speed control mode; the following decision trajectory point speed control mode is used to control the speed of the autonomous vehicle based on the expected speed of the autonomous vehicle at the trajectory point. Step 3: During the autonomous vehicle's operation within the second sub-region, if the deviation value is greater than or equal to the second preset value, the autonomous vehicle is determined to adopt the following decision trajectory point speed control mode; if the deviation value is less than the second preset value, the autonomous vehicle is determined to adopt the position following mode; the position following mode is used to control the speed of the autonomous vehicle based on the trajectory length between the autonomous vehicle and the stop point and the preset minimum controllable speed. Step 4: During the autonomous vehicle's operation within the first sub-region, if the deviation value is greater than or equal to the first preset value, determine that the autonomous vehicle adopts the following decision trajectory point speed control mode; if the deviation value is less than the first preset value, determine that the autonomous vehicle adopts the open-loop deceleration mode; the open-loop deceleration mode is used to control the deceleration of the autonomous vehicle based on real-time vehicle speed, distance tolerance, and road slope.

[0034] In this embodiment, differentiated judgment rules are set for the third sub-region, the second sub-region, and the first sub-region: the third sub-region defaults to the speed control mode following the decision trajectory point; the second and first sub-regions switch between the speed control mode following the decision trajectory point, the position following mode, and the open-loop deceleration mode based on the relationship between the deviation value and the corresponding preset value, ensuring the accuracy and adaptability of speed control in different scenarios. The speed control mode can refer to a specific control method in autonomous vehicle parking control that adapts to different parking control zones and driving states for precise speed adjustment, including at least one of the following: speed control mode following the decision trajectory point, position following mode, and open-loop deceleration mode. The speed control mode following the decision trajectory point can use the desired speed of each trajectory point on the autonomous vehicle's decision trajectory as the control target, adjusting the real-time speed of the autonomous vehicle to accurately track the desired speed, ensuring the coordination between speed and trajectory. In this mode, vehicle speed control is a fine-grained speed tracking control at the trajectory point level. It first generates a decision trajectory containing multiple continuous trajectory points based on the mining truck's current driving position, stop location, and planned parking trajectory. Each trajectory point is matched with a desired speed adapted to the parking stage (as the mining truck approaches the stop point, the desired speed at each trajectory point decreases in a step-like or smooth manner). Then, the desired speed at that trajectory point is used as the control target to achieve speed following. Position following mode can refer to using the real-time trajectory length between the unmanned vehicle and the stop point and the preset minimum controllable speed as the core basis, dynamically adjusting the speed to ensure the unmanned vehicle approaches the stop point along the planned trajectory while stabilizing the speed within the minimum controllable speed range. Open-loop deceleration mode can refer to determining deceleration parameters based on the unmanned vehicle's real-time speed, distance tolerance, and road gradient, and executing deceleration operations according to these parameters. The first and second preset values ​​are pre-set deviation thresholds used to distinguish the degree of deviation in different driving states. The first preset value is adapted to the first sub-region range, and the second preset value is adapted to the second sub-region range. The thresholds can be adjusted according to scenario requirements. The minimum controllable speed refers to the lowest speed at which the autonomous vehicle can drive stably without losing stability, providing a speed control benchmark for the position following mode. Distance tolerance can refer to the maximum allowable deviation between the final stopping position of the autonomous vehicle and the target stopping point (usually 10-15cm), providing a basis for deceleration accuracy in the open-loop deceleration mode. Regarding step one above, the deviation value between the expected driving state and the current driving state can refer to the comprehensive quantitative difference between the expected driving state (expected speed, expected trajectory length) calculated based on the expected driving state and the current driving state of the autonomous vehicle (such as real-time speed, real-time trajectory length). This is used to accurately determine whether the speed control mode needs to be switched, objectively reflecting the degree of deviation and the scope of influence between the current driving state and the planned target.Real-time data collection of the autonomous vehicle's current driving information, such as real-time speed and the trajectory length between the vehicle and the stopping point, is compared with a preset expected driving state (expected speed and expected trajectory length for the corresponding trajectory point). Using the expected driving state as a benchmark, an algorithm calculates the comprehensive quantitative difference (i.e., deviation value) between the two, accurately quantifying the degree of deviation between the current driving state and the planned target. For step two above, if the autonomous vehicle is currently in the third sub-region, there is no need to compare the deviation value; the following decision trajectory point speed control mode is directly adopted, and the speed is adjusted according to the control logic of this mode. The third sub-region represents the long-distance parking stage, where the goal is to enable the autonomous vehicle to quickly approach the planned trajectory and smoothly decelerate, without requiring precise alignment. The following decision trajectory point speed control mode can achieve efficient speed control and trajectory calibration in the long-distance stage by tracking the expected speed of the trajectory point, adapting to the control requirements of this area. For step three above, if it is determined that the autonomous vehicle is currently in the second sub-region, the deviation value is compared with a second preset value, and the corresponding speed control mode is switched based on the comparison result. If the deviation value is greater than or equal to the second preset value, it indicates that the overall difference between the current driving state (real-time speed and real-time trajectory length) and the expected driving state of the autonomous vehicle is within a reasonable range, and the real-time speed is still within a controllable range. For example, if the real-time speed is less than the expected speed, the following decision trajectory point speed control mode is adopted to continuously track the expected speed and drive smoothly to ensure that it approaches the planned trajectory. If the deviation value is less than the second preset value, it indicates that the current driving state of the autonomous vehicle deviates from the expected driving state, specifically, the expected speed is less than the real-time speed, or the expected trajectory length is greater than the real-time trajectory length. In this case, it is necessary to promptly reduce speed to return to the planned target. The position following mode is adopted, and based on the real-time trajectory length between the autonomous vehicle and the stop point and the preset minimum controllable speed, the speed is precisely adjusted to achieve speed reduction and ensure that the driving state conforms to the expected target. For step four above, if it is determined that the autonomous vehicle is currently in the first sub-region, the deviation value is compared with the first preset value, and the corresponding speed control mode is switched according to the comparison result. If the deviation value is greater than or equal to the first preset value, it indicates that the overall difference between the current driving state (real-time vehicle speed, real-time trajectory length) and the expected driving state of the autonomous vehicle is within a reasonable range, and the real-time vehicle speed is still within a controllable range. For example, if the real-time vehicle speed is less than the expected vehicle speed, the vehicle speed control mode following the decision trajectory point is adopted to continuously track the expected vehicle speed and reduce speed to stop. If the deviation value is less than the first preset value, it indicates that the current driving state of the autonomous vehicle deviates from the expected driving state. For example, if the real-time vehicle speed is greater than the expected vehicle speed, or the real-time trajectory length is less than the expected trajectory length, then timely and precise speed reduction adjustment is required to return to the planned target and ensure accurate stopping. The vehicle speed control mode is adopted, based on the real-time vehicle speed, distance tolerance, and road slope, to precisely control the deceleration of the autonomous vehicle and ensure that the driving state conforms to the expected target and the stopping accuracy meets the requirements.By accurately quantifying deviation values ​​and switching modes on demand, deviations in the autonomous vehicle's driving status can be identified in a timely manner. When the real-time vehicle speed is too high or the trajectory length is too short, deceleration control can be quickly initiated (position following mode in the second sub-region and open-loop deceleration mode in the first sub-region) to avoid speed fluctuations and vehicle impacts caused by accumulated deviations. When the status is controllable, the vehicle speed control mode following the decision trajectory point is maintained to ensure driving stability.

[0035] In another embodiment of this disclosure, in step one above, determining the deviation between the desired driving state and the current driving state of the autonomous vehicle based on the vehicle's driving information includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the autonomous vehicle is determined based on the real-time vehicle speed, the expected acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance. The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. The symbolic features of the sliding surface are extracted using a symbolic function to determine the deviation between the expected driving state and the current driving state of the unmanned vehicle.

[0036] In this embodiment, the deviation components in vehicle speed and position are combined with sliding mode control to achieve precise quantification of the deviation value. Driving information includes: time consumed for aiming at trajectory points, real-time vehicle speed, desired acceleration, vehicle speed control execution delay time, trajectory length between the autonomous vehicle and the stopping point, speed error weighting coefficient, and position error weighting coefficient. Based on the autonomous vehicle's driving information, the deviation value between the desired driving state and the current driving state is determined, expressed by the formula: ; in, This represents the deviation between the expected driving state and the current driving state of the autonomous vehicle. Represents a symbolic function. This represents the speed error weighting coefficient. This represents the position error weighting coefficient. Indicates the desired speed. This indicates the time consumed by the pre-aiming trajectory points. S represents the real-time vehicle speed, and S represents the trajectory length between the autonomous vehicle and the stop point. Indicates the expected acceleration. This indicates the delay time for vehicle speed control execution. This indicates the expected vehicle speed after the time consumed by the pre-aimed trajectory point; Indicates the velocity deviation component; This indicates the predicted remaining driving distance; This represents the positional deviation component. This indicates the sliding surface.

[0037] The deviation between the expected and current driving states of the autonomous vehicle (VV) is a comprehensive quantitative index obtained by weighting multi-dimensional deviation components and extracting the sliding surface symbol, based on the expected driving state. This objectively reflects the degree of deviation between the expected and current driving states and serves as the core criterion for subsequent speed control mode switching. The expected speed after the time consumed by the pre-aimed trajectory point is the target speed the VV should reach after the time consumed by the pre-aimed trajectory point on the planned trajectory, and is the core reference value for speed planning. The speed deviation component is the difference between the expected speed after the time consumed by the pre-aimed trajectory point and the VV's real-time speed. It quantitatively reflects the deviation between the current speed and the planned speed and is a core component of the deviation value in terms of speed dimensions. The speed control execution delay time is the time difference between the VV's speed control actuator issuing the speed control command and the execution mechanism (braking, power system) completing the action. This is an inherent characteristic of VV control and must be considered in deviation judgment. The predicted remaining travel distance is calculated by combining the autonomous vehicle's real-time speed, expected acceleration, and speed control execution delay time. The algorithm predicts the distance the autonomous vehicle will travel during the delay period, representing the potential risk of positional deviation caused by speed control lag. The positional deviation component is the difference between the real-time trajectory length of the autonomous vehicle and the parking point and the predicted remaining travel distance. It quantifies the deviation between the autonomous vehicle's current position and its planned position and is a core positional dimension component of the deviation value. The speed error weighting coefficient and position error weighting coefficient are pre-set weighting coefficients based on the autonomous vehicle's parking control stage (different sub-areas) and operating conditions. They are used to differentiate the importance of the speed deviation component and position deviation component in the overall deviation judgment and can be dynamically adjusted as needed. The sliding surface is a comprehensive quantitative surface obtained by weighting the speed error weighting coefficient, speed deviation component, position error weighting coefficient, and position deviation component. It is the intermediate core indicator connecting the multi-dimensional deviation components and the final deviation value. By considering both speed and position deviations and incorporating speed control execution delay into the deviation assessment system, and correcting for position deviation by predicting remaining travel distance, this effectively overcomes the distortion caused by traditional single-speed deviation assessments that ignore execution delays. It quantifies the deviation between vehicle speed and planned values ​​while also taking into account the risk of position overshoot due to execution lag, ensuring that the deviation value objectively and comprehensively reflects the actual driving state of the autonomous vehicle. This provides precise data support for the scientific switching of subsequent speed control modes, improving the accuracy of parking control from the source.

[0038] In another embodiment of this disclosure, when the autonomous vehicle is within a third sub-region and while driving within a second sub-region, after determining that the autonomous vehicle adopts a speed control mode following a decision trajectory point, the driving state of the autonomous vehicle is controlled according to the speed control mode, including: Step (1): Adopt the vehicle speed control mode of following the decision trajectory point, and control the unmanned vehicle to follow the decision trajectory point at the first vehicle speed; the first vehicle speed is not less than the preset minimum controllable vehicle speed; During the autonomous vehicle's operation within the first sub-region, after determining the vehicle's following decision trajectory point speed control mode, the vehicle's driving state is controlled according to the speed control mode, including: Step (1): Determine the first target acceleration of the unmanned vehicle based on its driving information; the driving information includes: the trajectory length between the unmanned vehicle and the stopping point, the real-time vehicle speed, and the road gradient; Step (2): Based on the first target acceleration, control the unmanned vehicle to decelerate and stop.

[0039] In this embodiment of the disclosure, if the autonomous vehicle is located in the third sub-region or is traveling within the second sub-region and it is determined to adopt the speed control mode of following the decision trajectory point, the autonomous vehicle is controlled to follow the first speed of the decision trajectory point, ensuring that the first speed is not lower than the preset minimum controllable speed; if the autonomous vehicle is traveling within the first sub-region and it is determined to adopt this mode, a first target acceleration is first determined based on the autonomous vehicle's driving information (trajectory length, real-time speed, road slope), and then the autonomous vehicle is controlled to decelerate and stop according to this acceleration, providing reliable execution support for the phased parking control of the autonomous vehicle. Figure 2 As shown, for step (1) above, following the decision trajectory point speed control mode, based on the current driving position of the mining truck, the stopping point position, and the planned parking trajectory, a decision trajectory containing multiple continuous trajectory points is generated. Each trajectory point is matched with a desired speed suitable for the parking stage. For example, as Figure 1As shown, between d1 and d2, the autonomous vehicle is in the third sub-region. Due to obstacles or digging, the vehicle's real-time speed may fall below the minimum controllable speed. If the planned speed is also lower than the minimum controllable speed, the first speed is set as the minimum controllable speed to ensure the vehicle's real-time speed returns to above it. Between d2 and d3, the vehicle enters the second sub-region. If the expected speed is greater than or equal to the minimum controllable speed, the expected speed is set as the first speed. This ensures stable vehicle operation and prevents instability (such as rolling away or stalling), thus guaranteeing the vehicle's stability. For step one above, the real-time data collected includes the vehicle's trajectory length from the parking point, real-time speed, and road gradient. The trajectory length determines the total deceleration distance, the real-time vehicle speed determines the initial deceleration baseline, and the road gradient is used to compensate for the impact of the gradient on the deceleration effect (e.g., reducing deceleration force when going uphill and increasing deceleration force when going downhill). These three factors work together to ensure the accuracy of the first target acceleration, meeting the stringent requirements of close-range stopping. The core of the first target acceleration is precise deceleration acceleration, the magnitude of which is determined by the trajectory length and the real-time vehicle speed (the shorter the trajectory length and the higher the real-time vehicle speed, the greater the absolute value of the deceleration acceleration). Simultaneously, dynamic compensation is applied based on the road gradient to ensure a smooth, shock-free deceleration process and precise stopping when the trajectory length is exhausted. The theoretical deceleration acceleration can be calculated based on kinematic formulas, combined with the trajectory length (remaining deceleration distance) and the real-time vehicle speed (initial speed), and then compensated and corrected using the road gradient coefficient to finally output the first target acceleration, ensuring that the acceleration adapts to the current vehicle state and road conditions. For step (ii) above, the first target acceleration is converted into control commands (brake pressure adjustment commands, power cut-off commands, etc.) that can be recognized by the unmanned vehicle's actuators and sent to the braking and power systems to control the unmanned vehicle to decelerate at a constant speed according to the first target acceleration. The real-time speed and trajectory length of the unmanned vehicle are monitored in real time until the unmanned vehicle accurately stops at the target position and completes the parking action.

[0040] In another embodiment of this disclosure, after determining that the autonomous vehicle is using an open-loop deceleration mode while it is driving within the first sub-region, the driving state of the autonomous vehicle is controlled according to the vehicle speed control mode, including: Based on the autonomous vehicle's driving information, determine the autonomous vehicle's second target acceleration, including: The vehicle speed component is determined by the vehicle speed coefficient and the square of the real-time vehicle speed. The distance components are determined based on the reciprocal of the sum of the trajectory length and the distance tolerance, and the distance coefficient. The slope components are determined based on the sine of the road slope, gravitational acceleration, and slope coefficient. The second target acceleration of the unmanned vehicle is determined based on the vehicle speed component, distance component, and slope component. Based on the second target acceleration, the driverless car is controlled to decelerate and stop.

[0041] In this embodiment, based on the autonomous vehicle's driving information, the vehicle speed component, distance component, and slope component are calculated in separate dimensions, and the three components are fused to obtain a second target acceleration. Then, the autonomous vehicle is controlled to perform open-loop deceleration based on the second target acceleration until a precise stop is achieved. The driving information includes: vehicle speed coefficient, real-time vehicle speed, trajectory length between the autonomous vehicle and the stopping point, distance tolerance, distance coefficient, road slope, and slope coefficient. The second target acceleration of the autonomous vehicle is determined based on its driving information, expressed by the formula: ; in, Indicates the acceleration of the second target. Indicates the vehicle speed coefficient. Indicates real-time vehicle speed. Represents the distance coefficient. This indicates the length of the trajectory between the autonomous vehicle and the stop point. Indicates distance tolerance. Indicates the slope coefficient. Represents gravitational acceleration. Indicates the road slope. This represents the vehicle speed component, which is the square of the real-time vehicle speed. The larger the real-time vehicle speed, the larger the absolute value of the vehicle speed component, and the stronger the deceleration force required. It is also the higher the vehicle speed, and the greater the actual control requirement for rapid deceleration in close-range situations. Representing the distance component, the shorter the trajectory length, the smaller the value of "trajectory length + distance tolerance" and the larger the reciprocal, the larger the absolute value of the distance component, which corresponds to a stronger deceleration force required, in line with the actual control requirement that "the closer the remaining distance, the greater the deceleration required to ensure precise stopping". The slope component represents the acceleration component. When going uphill, the sine of the road slope is positive, and the slope component is positive, which can offset some of the deceleration (going uphill itself has a deceleration effect). When going downhill, the sine of the road slope is negative, and the slope component is negative, which can increase the deceleration (going downhill has a coasting acceleration effect), achieving precise compensation for the deceleration effect of the slope. The vehicle speed component is an acceleration component calculated based on the real-time speed of the autonomous vehicle, reflecting the impact of real-time speed on the deceleration force, and is the core speed dimension basis for the second target acceleration. The distance component is an acceleration component calculated based on the remaining trajectory length and distance tolerance of the autonomous vehicle, reflecting the impact of the remaining stopping distance on the deceleration force, and is the core distance dimension basis for the second target acceleration. The slope component is an acceleration component calculated based on the road slope, used to compensate for the impact of road slope on the deceleration effect, and is the core operating condition dimension basis for the second target acceleration. The speed coefficient, distance coefficient, and gradient coefficient are pre-set proportional coefficients based on the performance of the unmanned vehicle (inertia, braking characteristics) and the operational scenario. They are used to quantify the influence of each dimension component on the acceleration of the second target and can be statically adjusted as needed. For example, =1、 =0.25、 =0.5, then as shown in Table 1 below, Table 1 is a quick lookup table for vehicle speed component, distance component, and gradient component.

[0042]

[0043] Table 1 The second target acceleration is converted into control commands (such as brake pressure setpoint commands) recognizable by the autonomous vehicle's braking system and sent to the actuators. This controls the autonomous vehicle to perform open-loop deceleration according to the second target acceleration until it precisely stops at the target position, completing the parking maneuver. Components are calculated separately for vehicle speed, distance, and gradient, and then integrated to achieve comprehensive quantification of deceleration requirements. The vehicle speed component matches the real-time vehicle speed reduction requirement, the distance component matches the parking accuracy requirement based on the remaining distance, and the gradient component compensates for the influence of road conditions. Each component calculation is tailored to the characteristics of close-range parking for autonomous vehicles, ensuring that the second target acceleration is adapted to both the real-time vehicle status and actual road conditions. This improves the accuracy of deceleration control from the source, avoiding deceleration inaccuracies caused by a single parameter. It ensures that the deviation between the final parking position and the parking point is within the allowable range (distance tolerance), stably achieving high-precision parking and meeting the stringent accuracy requirements of scenarios such as mine unloading and industrial park parking.

[0044] In another embodiment of this disclosure, after determining that the autonomous vehicle is in position-following mode while it is driving within the second sub-region, the driving state of the autonomous vehicle is controlled according to the vehicle speed control mode, including: Step 1) Determine the acceleration feedforward of the unmanned vehicle based on the deviation value; Step 2) The acceleration feedforward and the desired acceleration of the autonomous vehicle are superimposed to control the autonomous vehicle to travel at a second speed that is not less than the preset minimum controllable speed, and the difference between the second speed and the preset minimum controllable speed is within the first preset threshold range.

[0045] In this embodiment, an acceleration feedforward is calculated based on the deviation between the desired driving state and the current driving state of the autonomous vehicle, to compensate for driving state fluctuations caused by the deviation. The acceleration feedforward is then superimposed on the desired acceleration of the autonomous vehicle. Based on the superimposed acceleration, the autonomous vehicle is controlled to travel at a second speed, while strictly ensuring that the second speed is not less than a preset minimum controllable speed, and that the difference between the two is controlled within a first preset threshold range, ensuring that the autonomous vehicle smoothly approaches the parking point. This transforms the random initial state of the final stage of parking at the parking point into a fixed initial state input, improving parking accuracy and parking feel. Regarding step 1), the acceleration feedforward of the autonomous vehicle is determined based on the deviation value. For example, based on the current deviation value of the autonomous vehicle (less than a second preset value), combined with a preset feedforward coefficient, a standardized algorithm is used to calculate the acceleration feedforward used to compensate for deviation fluctuations, providing a compensation basis for subsequent acceleration superposition and speed control. The acceleration feedforward obtained in step 2) is superimposed with the desired acceleration to obtain the final executed acceleration. The desired acceleration can be a reference acceleration associated with the vehicle speed control mode of the following decision trajectory point, derived from the desired vehicle speed of the decision trajectory point in the current second sub-region of the autonomous vehicle. Based on the executed acceleration, a speed control command is issued to control the autonomous vehicle's movement. The actual driving speed of the autonomous vehicle (i.e., the second speed) is monitored in real time to ensure that the second speed meets the following constraints: not less than the minimum controllable speed. If the second speed is less than the minimum controllable speed, the minimum controllable speed is used as the second speed. The difference between the second speed and the minimum controllable speed does not exceed a first preset threshold until the autonomous vehicle leaves the second sub-region, enters the first sub-region, or switches to another speed control mode. The acceleration feedforward is calculated by the deviation value to compensate for small deviation fluctuations in the second sub-region in advance, avoiding speed fluctuations and position deviations caused by the accumulation of small deviations; the speed control accuracy is further optimized by the minimum controllable speed constraint to ensure that the second speed is stable within the adaptation range, allowing the unmanned vehicle to drive smoothly in the position following mode and improving the accuracy of position following.

[0046] In another embodiment of this disclosure, step 1) above, determining the acceleration feedforward of the unmanned vehicle based on the deviation value, includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the autonomous vehicle is determined based on the real-time vehicle speed, the expected acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance. The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. By extracting the sign features of the sliding surface using a sign function, the deviation between the expected driving state and the current driving state of the autonomous vehicle can be determined. The adaptive coefficient of the sliding surface is determined based on the deviation value, the preset adaptive coefficient, and the vehicle speed control execution cycle. The acceleration feedforward of the unmanned vehicle is determined based on the deviation value and the sliding surface adaptive coefficient.

[0047] In this embodiment, based on the deviations in vehicle speed and position, and combining sliding mode control with adaptive compensation logic, the speed and position deviation components are calculated step by step, weighted to obtain the sliding surface, and discrete deviation values ​​are extracted. Then, based on the deviation values ​​and preset parameters, the adaptive coefficient of the sliding surface is calculated. Finally, through the coordinated derivation of the deviation values ​​and the adaptive coefficient of the sliding surface, a precise acceleration feedforward is obtained. The acceleration feedforward of the autonomous vehicle is determined based on the deviation values, expressed by the formula: ; ; ; in, Indicates the acceleration feedforward quantity. This represents the adaptive coefficient of the sliding surface. This represents the adaptive coefficient of the sliding surface. Represents the adaptive coefficient. This indicates the vehicle speed control execution cycle. The vehicle speed control execution cycle refers to the fixed time interval between the autonomous vehicle control system issuing a speed control command and the actuator completing one action. It serves as the time dimension for quantifying the control rhythm and deriving the sliding surface adaptive coefficient. The formula " The core idea of ​​";" is to dynamically adjust the adaptive coefficient of the sliding surface based on the magnitude and direction of the current sliding surface (comprehensive deviation). This allows the compensation level to adapt to deviations in real time, rather than being a fixed value. Continuous sliding surfaces Converting the values ​​to discrete values, such as ±1 / 0, significantly simplifies the calculation logic. This eliminates the need for complex continuous value operations; only sign recognition is required to determine the compensation direction, improving the response speed of speed control commands. The sliding surface adaptive coefficient dynamically changes with the deviation value, combining a preset compensation benchmark with the real-time deviation state of the autonomous vehicle. This allows the final acceleration feedforward to precisely adjust the compensation intensity based on the deviation trend. Through precise feedforward compensation, small deviations within the second sub-region can be quickly corrected, avoiding speed fluctuations and position shifts caused by accumulated deviations. This standardizes the autonomous vehicle's driving state, providing a smooth and unified initial state for subsequent entry into the first sub-region and switching to the near-distance precise stopping mode, effectively bridging the long-distance deceleration and near-distance stopping phases.

[0048] In another embodiment of this disclosure, the method further includes: when the location of the stop point represented by the trajectory length information changes, switching the vehicle speed control mode according to the driving information, and controlling the driving state of the unmanned vehicle according to the switched vehicle speed control mode. The vehicle speed control mode is switched based on driving information, including one or a combination of the following: Case 1: If the current vehicle speed control mode is the location following mode and the driving information meets the first condition, switch the current vehicle speed control mode to the following decision trajectory point speed control mode; the first condition includes: the trajectory length is greater than the first preset distance threshold; or, the trajectory length is between the second preset distance threshold and the first preset distance threshold, and the duration exceeds the preset time threshold; or, the real-time vehicle speed is less than or equal to the first vehicle speed threshold; the first preset distance threshold is greater than the second preset distance threshold. Case 2: If the current vehicle speed control mode is position following mode and the driving information meets the second condition, switch the current vehicle speed control mode to open-loop deceleration mode; the second condition includes: the trajectory length is less than the third preset distance threshold and the deviation value is less than the first preset value; or, the trajectory length is less than the fourth preset distance threshold; the third preset distance threshold is greater than the fourth preset distance threshold. Case 3: If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the third condition, the current vehicle speed control mode will be switched to the position following mode. The third condition includes: the trajectory length is between the third preset distance threshold and the second preset distance threshold, the real-time vehicle speed is greater than the first vehicle speed threshold, the driverless vehicle is not in the starting state, the deviation value is less than the second preset value, and the current expected acceleration is less than the preset acceleration threshold. Case 4: If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the fourth condition, switch the current vehicle speed control mode to the open-loop deceleration mode; the fourth condition includes: the trajectory length is less than the third preset distance threshold, the driverless vehicle is not in the starting state, the deviation value is less than the first preset value, and the real-time vehicle speed is between the first vehicle speed threshold and the second vehicle speed threshold; or, the trajectory length is less than the fourth preset distance threshold, and the real-time vehicle speed is greater than the first vehicle speed threshold. Case 5: If the current vehicle speed control mode is open-loop deceleration mode and the driving information meets the fifth condition, switch the current vehicle speed control mode to the following decision trajectory point speed control mode; the fifth condition includes: the real-time vehicle speed is less than or equal to the first vehicle speed threshold, or the trajectory length is greater than the second preset distance threshold.

[0049] In this embodiment, when the autonomous vehicle detects a change in the parking point location, it dynamically switches the vehicle speed control mode. The parking point location can change; for example, the autonomous vehicle may use the rear position of another autonomous vehicle as its parking point. During parking control, if the other autonomous vehicle moves, the parking point location changes. Similarly, in an unloading scenario, if the unloading point changes during parking control, the autonomous vehicle's parking point location changes. When the trajectory length information indicates a change in the parking point location, the vehicle speed control mode is switched based on the driving information. The autonomous vehicle's driving state is then controlled according to the switched speed control mode, resolving the problem of mismatch between the original speed control mode and the parking point location. The autonomous vehicle collects driving information in real time and matches it with the corresponding switching judgment rule based on the current speed control mode to determine whether the driving information meets the switching conditions for the speed control mode. If it does, the mode switch is completed, and the autonomous vehicle's driving state is finally controlled according to the switched speed control mode. The driving information includes: the trajectory length between the autonomous vehicle and the parking point, the real-time speed, whether the autonomous vehicle is in a starting state, and the current desired acceleration. The vehicle speed control mode, position following mode, and open-loop deceleration mode following the decision trajectory point can be represented by a state machine, and the switching of vehicle speed control mode corresponds to the transition of the state machine. The first preset distance threshold, the second preset distance threshold, the third preset distance threshold, and the fourth preset distance threshold can be the trajectory length critical values ​​preset according to the needs of each stage of autonomous vehicle parking control. They satisfy the relationship that the first preset distance threshold > the second preset distance threshold > the third preset distance threshold > the fourth preset distance threshold, which is the judgment criterion for switching vehicle speed control modes in the distance dimension.

[0050] like Figure 3As shown, for situation 1 above, S represents the trajectory length, and the current vehicle speed control mode is position following mode 301. If the trajectory length is greater than the first preset distance threshold, which can be represented by d1, i.e., S > d1, then switch to following decision trajectory point speed control mode 302. If the trajectory length is between the second preset distance threshold and the first preset distance threshold, and the duration exceeds the preset time threshold, which can be represented by d2, then if d1 ≥ S > d2, i.e., the autonomous vehicle is in the third sub-region, T_P2N represents the counter, which increments by 1 every algorithm cycle (0.02s), accumulating 50 cycles to 1s (if d1 ≥ S > d2, T_P2N++), and the preset time threshold can be 1s. If T_P2N > 50 (1s), then switch to following decision trajectory point speed control mode 302. The real-time vehicle speed is represented by VehSpd. The real-time vehicle speed is less than or equal to the first vehicle speed threshold, which can be 0.1 m / s. If VehSpd ≤ 0.1 m / s, it means that the vehicle is about to stop, and then it switches to the initial mode, that is, switches to the speed control mode 302 that follows the decision trajectory point.

[0051] Regarding scenario 2 above, if the current vehicle speed control mode is position following mode 301, and the driving information meets the second condition, the current vehicle speed control mode is switched to open-loop deceleration mode 303. The second condition includes: the trajectory length is less than a third preset distance threshold, and the deviation value is less than a first preset value. The third preset distance threshold can be d3, i.e., S < d3, indicating that the autonomous vehicle is within the first sub-region, and the deviation value is less than the first preset value, thus switching to open-loop deceleration mode 303. The deviation value is represented as sign(s), and the first preset value can be 0, i.e., sign(s) < 0. Alternatively, the trajectory length is less than a fourth preset distance threshold, indicating that the autonomous vehicle is suddenly very close to the stopping point. The fourth preset distance threshold can be 0.3m, i.e., S < 0.3m, thus switching to open-loop deceleration mode 303 to quickly control the autonomous vehicle to stop.

[0052] Regarding situation 3 above, if the current vehicle speed control mode is the following decision trajectory point speed control mode 302, and the driving information meets the third condition, the current vehicle speed control mode will be switched to the position following mode 301. The third condition includes: the trajectory length is between the third preset distance threshold and the second preset distance threshold, i.e., d3 < S < d2; the unmanned vehicle is within the second sub-region; the real-time vehicle speed is greater than the first vehicle speed threshold, VehSpd > 0.1 m / s; the unmanned vehicle is not in a starting state, represented as ! flg_drvoff; the deviation value is less than the second preset value, which can be 0, sign(s) < 0; and the current expected acceleration is less than the preset acceleration threshold, the expected acceleration is represented as dsrAx, and the preset acceleration threshold can be -0.3 m / s². 2 dsrAx < -0.3 m / s2 Since a smaller expected acceleration indicates a faster real-time vehicle speed, a more effective vehicle speed control mode is needed. Therefore, the vehicle is switched to position following mode 301.

[0053] Regarding situation 4 above, if the current vehicle speed control mode is the following decision trajectory point speed control mode 302, and the driving information meets the fourth condition, the current vehicle speed control mode will be switched to the open-loop deceleration mode 303. The fourth condition includes: the trajectory length is less than the third preset distance threshold, S < d3, indicating that the unmanned vehicle is within the first sub-region; the unmanned vehicle is not in a starting state, indicated by ! flg_drvoff; the deviation value is less than the first preset value (sign(s) < 0), and the real-time vehicle speed is between the first and second speed thresholds. The second speed threshold can be 1.11 m / s, expressed as 1.11 m / s > VehSpd > 0.1 m / s, where VehSpd > 0.1 m / s indicates that the unmanned vehicle has not yet stopped, and 1.11 m / s > VehSpd is to avoid the safety risk of sudden stopping from a large speed; or, the trajectory length is less than the fourth preset distance threshold, S < 0.3 m, and the real-time vehicle speed is greater than the first speed threshold (VehSpd > 0.1 m / s).

[0054] Regarding situation 5 above, if the current vehicle speed control mode is open-loop deceleration mode 303, and the driving information meets the fifth condition, the current vehicle speed control mode will be switched to following the decision trajectory point speed control mode 302. The fifth condition includes: the real-time vehicle speed is less than or equal to the first vehicle speed threshold (VehSpd≤0.1m / s), or the trajectory length is greater than the second preset distance threshold (S>d2), indicating that the unmanned vehicle is not within the first sub-region.

[0055] In another embodiment of this disclosure, the driving information includes the trajectory length between the unmanned vehicle and the stop point; In step S101 above, obtaining driving information of the unmanned vehicle within the preset area of ​​the parking point includes: The distance between the unmanned vehicle and the parking point is obtained using preset equipment, and this distance is used as the trajectory length between the unmanned vehicle and the parking point; the preset equipment includes: sensing equipment and / or positioning equipment; If the absolute value of the difference between the current trajectory length and the trajectory length at the previous moment is greater than the second preset threshold, the remaining length of the decision trajectory point will be used as the new trajectory length.

[0056] In this embodiment, the sensing device is a device that acquires distance data through environmental sensing technology, such as lidar, millimeter-wave radar, and visual cameras, which can directly detect the relative distance between the unmanned vehicle and the parking point. The positioning device can acquire the position coordinates of the unmanned vehicle and the parking point through spatial positioning technology, and then calculate the distance, such as GPS / BeiDou positioning modules, inertial navigation equipment, and indoor positioning base stations. The distance from the unmanned vehicle to the parking point is collected by the sensing device and / or the positioning device, and this distance is directly used as the trajectory length between the unmanned vehicle and the parking point. An abnormal trajectory length is determined by calculating the absolute value of the difference between the current trajectory length and the trajectory length at the previous moment. If this value is greater than a second preset threshold, the trajectory length acquisition is determined to be abnormal, indicating an abnormal change, such as the sensing device being blocked causing a sudden change in distance, or the positioning device losing a signal causing a coordinate shift. The remaining length of the decision trajectory point is used as the new trajectory length. If it does not exceed the threshold, the currently acquired trajectory length remains valid. The remaining length of the decision trajectory point refers to the remaining path length from the current decision trajectory point to the parking point in the preset decision trajectory for unmanned vehicle parking control. It is a correction replacement value when the trajectory length is abnormal and is a pre-planned fixed value. By comparing the absolute value of the difference between the current trajectory length and the trajectory length at the previous moment with the second preset threshold, abnormal changes in trajectory length caused by equipment failure, environmental obstruction, signal loss, etc. can be quickly identified. After the abnormality is determined, the remaining length of the pre-planned decision trajectory point is used for correction, replacing invalid collected data, avoiding problems such as incorrect switching of vehicle speed control mode and inaccurate speed control caused by abnormal data entering the subsequent control link, and ensuring that the trajectory length is always stable and effective.

[0057] Based on the same disclosed concept, this disclosure also provides a parking control device and an unmanned vehicle. Since the principle of solving the problem by these devices and unmanned vehicles is similar to that of the aforementioned parking control method, the implementation of the device and unmanned vehicle can refer to the implementation of the aforementioned method, and the repeated parts will not be described again.

[0058] This disclosure provides a parking control device, including a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the parking control device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the parking control method as described in any of the above embodiments are performed.

[0059] This disclosure provides an unmanned vehicle, including a parking control device as described in any of the above embodiments.

[0060] Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments of this disclosure can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this disclosure.

[0061] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes in the drawings are not necessarily essential for implementing this disclosure.

[0062] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0063] The sequence numbers of the embodiments disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0064] Obviously, those skilled in the art can make various modifications and variations to this disclosure without departing from its spirit and scope. Therefore, if such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, this disclosure is also intended to include such modifications and variations.

Claims

1. A parking control method, characterized in that, include: Obtain driving information of the unmanned vehicle when it reaches the preset area of ​​the parking point; Based on the driving information, the current parking control stage of the unmanned vehicle is determined; The driving state of the unmanned vehicle is controlled according to the parking control strategy corresponding to the parking control stage in which the unmanned vehicle is located, so as to complete the parking.

2. The method as described in claim 1, characterized in that, The driving information includes the trajectory length between the unmanned vehicle and the stop point; Based on the driving information, the current parking control phase of the autonomous vehicle is determined, including: The parking control zone where the unmanned vehicle is currently located is determined based on the trajectory length between the unmanned vehicle and the parking point; Based on the parking control zone and the parking control strategy corresponding to the parking control zone, the speed control mode of the driverless car is determined. The step of controlling the driving state of the autonomous vehicle according to the parking control strategy corresponding to the parking control stage in which the autonomous vehicle is located includes: The driving state of the unmanned vehicle is controlled according to the vehicle speed control mode.

3. The method as described in claim 2, characterized in that, The parking control zone is a continuous distance interval that pre-divides the preset area range; The parking control zone includes: a first sub-region, a second sub-region, and a third sub-region along the decision trajectory of the unmanned vehicle, starting from the parking point and extending from the parking point to the current position of the unmanned vehicle.

4. The method as described in claim 3, characterized in that, The vehicle speed control modes include: following decision trajectory point speed control mode, position following mode, and open-loop deceleration mode; The step of determining the vehicle speed control mode of the autonomous vehicle based on the parking control zone and the corresponding parking control strategy includes: Based on the driving information of the unmanned vehicle, determine the deviation between the expected driving state and the current driving state of the unmanned vehicle; When the unmanned vehicle is within the third sub-region, it is determined that the unmanned vehicle adopts the following decision trajectory point speed control mode; the following decision trajectory point speed control mode is used to control the speed of the unmanned vehicle based on the expected speed of the unmanned vehicle at the trajectory point. During the autonomous vehicle's operation within the second sub-region, if the deviation value is greater than or equal to a second preset value, the autonomous vehicle is determined to adopt a following decision trajectory point speed control mode; if the deviation value is less than the second preset value, the autonomous vehicle is determined to adopt a position following mode; the position following mode is used to control the speed of the autonomous vehicle based on the trajectory length between the autonomous vehicle and the stop point and a preset minimum controllable speed. During the autonomous vehicle's operation within the first sub-region, if the deviation value is greater than or equal to a first preset value, the autonomous vehicle is determined to adopt a speed control mode that follows the decision trajectory point; if the deviation value is less than the first preset value, the autonomous vehicle is determined to adopt an open-loop deceleration mode; the open-loop deceleration mode is used to control the deceleration of the autonomous vehicle based on real-time vehicle speed, distance tolerance, and road slope.

5. The method as described in claim 4, characterized in that, The step of determining the deviation between the expected driving state and the current driving state of the autonomous vehicle based on the driving information of the autonomous vehicle includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the unmanned vehicle is determined based on the real-time vehicle speed, the desired acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance; The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. The symbolic features of the sliding surface are extracted using a symbolic function to determine the deviation between the desired driving state and the current driving state of the unmanned vehicle.

6. The method as described in claim 4, characterized in that, When the autonomous vehicle is within the third sub-region, and while the autonomous vehicle is traveling within the second sub-region, after determining that the autonomous vehicle adopts a speed control mode following a decision trajectory point, controlling the driving state of the autonomous vehicle according to the speed control mode includes: The vehicle speed control mode following the decision trajectory point is adopted to control the unmanned vehicle to follow the decision trajectory point at a first speed; the first speed is not less than the preset minimum controllable speed. During the autonomous vehicle's operation within the first sub-region, after determining the following decision trajectory point speed control mode adopted by the autonomous vehicle, controlling the autonomous vehicle's driving state according to the speed control mode includes: Based on the driving information of the unmanned vehicle, the first target acceleration of the unmanned vehicle is determined; wherein, the driving information includes: the trajectory length between the unmanned vehicle and the stopping point, the real-time vehicle speed, and the road gradient; Based on the first target acceleration, the unmanned vehicle is controlled to decelerate and stop.

7. The method as described in claim 4, characterized in that, During the autonomous vehicle's operation within the first sub-region, after determining that the autonomous vehicle is using an open-loop deceleration mode, controlling the autonomous vehicle's driving state according to the vehicle speed control mode includes: Based on the driving information of the unmanned vehicle, the second target acceleration of the unmanned vehicle is determined, including: The vehicle speed component is determined by the vehicle speed coefficient and the square of the real-time vehicle speed. The distance component is determined based on the reciprocal of the sum of the trajectory length and the distance tolerance, and the distance coefficient; The slope components are determined based on the sine of the road slope, gravitational acceleration, and slope coefficient. The second target acceleration of the unmanned vehicle is determined based on the vehicle speed component, the distance component, and the slope component. Based on the second target acceleration, the unmanned vehicle is controlled to decelerate and stop.

8. The method as described in claim 4, characterized in that, During the autonomous vehicle's operation within the second sub-region, after determining that the autonomous vehicle is in position-following mode, controlling the autonomous vehicle's driving state according to the vehicle speed control mode includes: Based on the deviation value, determine the acceleration feedforward of the unmanned vehicle; The acceleration feedforward and the desired acceleration of the unmanned vehicle are superimposed to control the unmanned vehicle to travel at a second speed that is not less than a preset minimum controllable speed, and the difference between the second speed and the preset minimum controllable speed is within a first preset threshold range.

9. The method as described in claim 8, characterized in that, The step of determining the acceleration feedforward of the unmanned vehicle based on the deviation value includes: The speed deviation component is determined based on the difference between the expected vehicle speed after the time consumed by the pre-aimed trajectory point and the real-time vehicle speed. The predicted remaining driving distance of the unmanned vehicle is determined based on the real-time vehicle speed, the desired acceleration, and the vehicle speed control execution delay time. The position deviation component is determined based on the difference between the trajectory length and the predicted remaining travel distance; The sliding surface is determined by weighting the velocity error weighting coefficient, the velocity deviation component, the position error weighting coefficient, and the position deviation component. The sign features of the sliding surface are extracted by the sign function to determine the deviation between the expected driving state and the current driving state of the unmanned vehicle. The sliding surface adaptive coefficient is determined based on the deviation value, the preset adaptive coefficient, and the vehicle speed control execution cycle. The acceleration feedforward of the unmanned vehicle is determined based on the deviation value and the sliding surface adaptive coefficient.

10. The method as described in claim 4, characterized in that, The method further includes: when the location of the stop point represented by the trajectory length information changes, switching the vehicle speed control mode according to the driving information, and controlling the driving state of the unmanned vehicle according to the switched vehicle speed control mode; The step of switching the vehicle speed control mode based on the driving information includes one or a combination of the following: If the current vehicle speed control mode is the location following mode, and the driving information meets the first condition, the current vehicle speed control mode is switched to the following decision trajectory point speed control mode; the first condition includes: the trajectory length is greater than a first preset distance threshold; or, the trajectory length is between a second preset distance threshold and a first preset distance threshold, and the duration exceeds a preset time threshold; or, the real-time vehicle speed is less than or equal to a first vehicle speed threshold; the first preset distance threshold is greater than the second preset distance threshold; If the current vehicle speed control mode is position following mode, and the driving information meets the second condition, the current vehicle speed control mode is switched to open-loop deceleration mode; the second condition includes: the trajectory length is less than the third preset distance threshold, and the deviation value is less than the first preset value; or, the trajectory length is less than the fourth preset distance threshold; the third preset distance threshold is greater than the fourth preset distance threshold. If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the third condition, the current vehicle speed control mode is switched to the position following mode; the third condition includes: the trajectory length is between the third preset distance threshold and the second preset distance threshold, the real-time vehicle speed is greater than the first vehicle speed threshold, the unmanned vehicle is not in the starting state, the deviation value is less than the second preset value, and the current expected acceleration is less than the preset acceleration threshold. If the current vehicle speed control mode is the following decision trajectory point speed control mode, and the driving information meets the fourth condition, the current vehicle speed control mode is switched to the open-loop deceleration mode; the fourth condition includes: the trajectory length is less than the third preset distance threshold, the unmanned vehicle is not in a starting state, the deviation value is less than the first preset value, and the real-time vehicle speed is between the first vehicle speed threshold and the second vehicle speed threshold; or, the trajectory length is less than the fourth preset distance threshold, and the real-time vehicle speed is greater than the first vehicle speed threshold. If the current vehicle speed control mode is open-loop deceleration mode, and the driving information meets the fifth condition, the current vehicle speed control mode will be switched to the following decision trajectory point speed control mode; the fifth condition includes: the real-time vehicle speed is less than or equal to the first vehicle speed threshold, or the trajectory length is greater than the second preset distance threshold.

11. The method as described in claim 1, characterized in that, The driving information includes the trajectory length between the unmanned vehicle and the stop point; The acquisition of driving information of the unmanned vehicle within the preset area of ​​the parking point includes: The distance between the unmanned vehicle and the parking point is obtained using a preset device, and this distance is used as the trajectory length between the unmanned vehicle and the parking point; the preset device includes: a sensing device and / or a positioning device; If the absolute value of the difference between the current trajectory length and the trajectory length at the previous moment is greater than the second preset threshold, the remaining length of the decision trajectory point will be used as the new trajectory length.

12. A parking control device, characterized in that, include: The system includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the parking control device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the parking control method as described in any one of claims 1 to 11 are performed.

13. An unmanned vehicle, characterized in that, Includes the parking control device as described in claim 12.

Images