Railway vehicle automatic driving method and device and railway vehicle

Abstract

The invention provides a rail vehicle automatic driving method and device and a rail vehicle, the method is applied to the rail vehicle, the rail vehicle runs on a vehicle road, and the method comprises the following steps: constructing running constraints of the rail vehicle running on the vehicle road; based on the operation constraint, determining a reference speed curve of the rail vehicle running on the vehicle road; and based on the reference speed curve, performing automatic driving control on the rail vehicle through a double-loop longitudinal tracking control mode, the double-loop longitudinal tracking control mode comprises control over the rail vehicle in the automatic driving process based on speed information and control over the rail vehicle in the automatic driving process based on position information. The minimum energy consumption of the railway vehicle and the optimal experience of passengers are ensured in the automatic driving process.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous driving, and particularly to an autonomous driving method, device and rail vehicle for rail vehicles. Background Art

[0002] As an innovative public transportation mode, virtual rail trains, with their unique design of multiple carriages in series and the operating characteristics of not requiring physical tracks, have demonstrated great potential in solving urban traffic congestion problems. However, compared with trains such as subways and trains with physical track constraints, the autonomous driving technology of virtual rail trains faces more complex challenges.

[0003] Current autonomous driving methods are mainly applicable to trains such as subways and trains with relatively closed operating scenarios and fewer uncertainties during the driving process. In contrast, virtual rail trains need to mix with social vehicles, operate in a complex traffic environment and strictly abide by traffic rules. The uncertainty of this operating environment makes the autonomous driving problem of virtual rail trains more complex and challenging.

[0004] Therefore, finding an autonomous driving method for rail vehicles that can efficiently control the automatic driving of rail vehicles has become a current research hotspot. Summary of the Invention

[0005] The present invention provides an autonomous driving method, device and rail vehicle for rail vehicles, which realizes ensuring the minimum energy consumption of rail vehicles and the optimal experience of passengers during the automatic driving process.

[0006] The present invention provides an autonomous driving method for rail vehicles. The method is applied to a rail vehicle that travels on a vehicle road. The method includes: constructing an operation constraint for the rail vehicle to travel on the vehicle road; determining a reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraint; and performing autonomous driving control on the rail vehicle through a double-loop longitudinal tracking control method. The double-loop longitudinal tracking control method includes controlling the rail vehicle during the autonomous driving process based on speed information and controlling the rail vehicle during the autonomous driving process based on position information.

[0007] A method for automatic driving of a rail vehicle provided by the present invention, based on the reference speed curve, automatically driving and controlling the rail vehicle through a double-loop longitudinal tracking control method, specifically including: determining a reference position curve corresponding to the reference speed curve based on the reference speed curve; automatically driving and controlling the rail vehicle through a double-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, so that the deviation generated during the driving of the rail vehicle is less than or equal to a deviation threshold, wherein the deviation includes a deviation in the speed dimension and a deviation in the position dimension.

[0008] A method for automatic driving of a rail vehicle provided by the present invention, after automatically driving and controlling the rail vehicle through a double-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the method further includes: obtaining the instantaneous speed of the rail vehicle at the current moment, and / or the instantaneous position of the rail vehicle at the current moment; determining the deviation of the rail vehicle at the current moment based on the instantaneous speed and the reference speed curve, and / or based on the instantaneous position and the reference position curve; when the deviation is greater than the deviation threshold, obtaining the instantaneous operation information of the rail vehicle and the instantaneous road information of the vehicle road in real time, wherein the instantaneous operation information at least includes instantaneous acceleration information and instantaneous speed information; the instantaneous road information at least includes the phase information of the traffic light and the instantaneous speed limit information; re-optimizing the reference speed curve based on the instantaneous operation information and the instantaneous road information to obtain an optimized reference speed curve, and automatically driving and controlling the rail vehicle based on the optimized reference speed curve.

[0009] A method for automatic driving of a rail vehicle provided by the present invention, after automatically driving and controlling the rail vehicle through a double-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the method further includes: when it is detected that there is an obstacle in front of the rail vehicle, obtaining the first running speed of the rail vehicle and the second running speed of the obstacle; determining the running acceleration of the rail vehicle based on the first running speed and the second running speed; obtaining a temporary reference speed curve corresponding to the running acceleration based on the running acceleration; automatically driving and controlling the rail vehicle based on the temporary reference speed curve.

[0010] A method for automatic driving of a rail vehicle provided by the present invention, after automatically driving and controlling the rail vehicle based on the temporary reference speed curve, the method further includes: when it is detected that the obstacle has left, automatically driving and controlling the rail vehicle through a double-loop longitudinal tracking control method based on the reference speed curve.

[0011] A method for automatic driving of a rail vehicle provided by the present invention constructs the operation constraints in the following manner: determining the arrival operation time information of the rail vehicle in a road sub-interval of the vehicle road and the speed limit information of the rail vehicle in the road sub-interval; constructing the operation constraints based on the arrival operation time information and the speed limit information; determining the reference speed curve of the rail vehicle traveling on the vehicle road based on the operation constraints, specifically including: respectively determining a passenger comfort index function, an energy consumption index function, and a punctuality index function, and constructing an objective optimization function under the road sub-interval based on the passenger comfort index function, the energy consumption index function, and the punctuality index function, wherein the passenger comfort index function is determined according to the acceleration of the rail vehicle; the energy consumption index function is determined according to the traction force and / or braking force of the rail vehicle; the punctuality index function is determined according to the arrival operation time information of the rail vehicle in the road sub-interval; determining the sub-interval reference speed curve of the rail vehicle in the road sub-interval based on the objective optimization function and the operation constraints so that the function value of the objective optimization function is maximized under the sub-interval reference speed curve; and determining the reference speed curve of the rail vehicle traveling on the vehicle road based on the sub-interval reference speed curves of the rail vehicle in each road sub-interval.

[0012] A method for automatic driving of a rail vehicle provided by the present invention, the curve trend of the sub-interval reference speed curve is successively running at a constant speed according to the starting speed, accelerating from the starting speed to the maximum operable speed based on a first duration according to the target acceleration, maintaining a constant speed at the maximum operable speed, and decelerating from the maximum operable speed to the termination speed, wherein the starting speed is the running speed of the rail vehicle at the starting position of the road sub-interval; the termination speed is the running speed of the rail vehicle at the termination position of the road sub-interval; the sub-interval reference speed curve is determined in the following manner: determining the target acceleration and the first duration; and determining the sub-interval reference speed curve according to the curve trend of the sub-interval reference speed curve in combination with the target acceleration and the first duration.

[0013] A method for automatic driving of a rail vehicle provided by the present invention, the target acceleration and the first duration are determined in the following manner: sampling multiple groups of candidate target accelerations and candidate first durations at a sampling interval, and based on the multiple groups of candidate target accelerations and candidate first durations, obtaining multiple candidate sub-interval reference speed curves corresponding to the multiple groups of candidate target accelerations and candidate first durations, and multiple candidate transfer costs corresponding to the multiple candidate sub-interval reference speed curves; obtaining the minimum candidate transfer cost based on the multiple candidate transfer costs; and determining the target acceleration and the first duration based on the candidate sub-interval reference speed curve corresponding to the minimum candidate transfer cost.

[0014] The present invention also provides an automatic driving device for a rail vehicle. The device is applied to a rail vehicle that travels on a vehicle road. The device includes: a construction module for constructing the operation constraints for the rail vehicle to travel on the vehicle road; a determination module for determining the reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraints; and a control module for performing automatic driving control on the rail vehicle by means of a double-loop longitudinal tracking control method based on the reference speed curve, where the double-loop longitudinal tracking control method includes control of the rail vehicle during the automatic driving process based on speed information and control of the rail vehicle during the automatic driving process based on position information.

[0015] The present invention also provides a rail vehicle, which includes: a rail vehicle body and a processor, where the processor is used to execute any one of the rail vehicle automatic driving methods.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements any one of the above-mentioned rail vehicle automatic driving methods.

[0017] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements any one of the above-mentioned rail vehicle automatic driving methods.

[0018] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements any one of the above-mentioned rail vehicle automatic driving methods.

[0019] The present invention provides an automatic driving method, device and rail vehicle for a rail vehicle running on a vehicle road. The method includes: constructing running constraints for the rail vehicle running on the vehicle road; determining a reference speed curve for the rail vehicle running on the vehicle road based on the running constraints; and performing automatic driving control on the rail vehicle through a double-loop longitudinal tracking control method based on the reference speed curve, so as to ensure the minimum energy consumption of the rail vehicle and the optimal experience of passengers during the automatic driving process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 is one of the flow diagrams of the automatic driving method for the rail vehicle provided by the present invention.

[0022] Figure 2 is another flow diagram of the automatic driving method for the rail vehicle provided by the present invention.

[0023] Figure 3 is yet another flow diagram of the automatic driving method for the rail vehicle provided by the present invention.

[0024] Figure 4 is the flow diagram for determining the reference speed curve for the rail vehicle running on the vehicle road based on the running constraints provided by the present invention.

[0025] Figure 5 is a schematic diagram of the reference speed curve for a sub-interval provided by the present invention.

[0026] Figure 6 is the structural diagram of the automatic driving device for the rail vehicle provided by the present invention.

[0027] Figure 7 is the structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments in the present invention fall within the scope of protection of the present invention.

[0029] The automatic driving method for rail vehicles provided by the present invention, based on vehicle-road cooperation technology, plans the driving speed curve of the train between adjacent stations (corresponding to the reference speed curve) by considering train punctual operation constraints, road condition constraints, and vehicle dynamics constraints (corresponding to operation constraints), so as to ensure performance indicators such as train riding comfort, punctuality, and economy, and realizes the automatic driving of virtual rail trains through a scenario-adaptive longitudinal controller.

[0030] Figure 1 It is one of the schematic flowcharts of the automatic driving method for rail vehicles provided by the present invention.

[0031] The following will be combined with Figure 1 to describe the process of the automatic driving method for rail vehicles provided by the present invention.

[0032] In an exemplary embodiment of the present invention, the automatic driving method for rail vehicles can be applied to rail vehicles. Among them, the rail vehicle can be a virtual rail train, and the specific form of the rail vehicle is not limited in this embodiment. During the application process, the rail vehicle can travel on a vehicle road. Combining Figure 1 it can be known that the automatic driving method for rail vehicles can include steps 110 to 130, and each step will be introduced separately below.

[0033] In step 110, the operation constraints for the rail vehicle to travel on the vehicle road are constructed.

[0034] In one embodiment, the operation constraints for the rail vehicle to travel on the vehicle road can be constructed according to the characteristics of the rail vehicle, the conditions of the vehicle road, and relevant traffic rules and safety standards. These operation constraints include but are not limited to: maximum allowable speed, minimum safety distance, emergency braking distance, speed limit on curves, signal light status, etc.

[0035] In step 120, based on the operation constraints, the reference speed curve for the rail vehicle to travel on the vehicle road is determined.

[0036] In one embodiment, after the operation constraints are constructed, the reference speed curve for the rail vehicle to travel on the vehicle road can be determined based on these constraints. The reference speed curve can be a curve representing the ideal speed values that the vehicle should reach at different positions. It comprehensively considers the dynamic characteristics of the rail vehicle, road conditions, and safety constraints to ensure that the vehicle can travel in an efficient and safe manner, so as to ensure the minimum energy consumption of the rail vehicle and the optimal experience for passengers during the automatic driving process.

[0037] In step 130, based on the reference speed curve, the rail vehicle is controlled for autonomous driving through a dual-loop longitudinal tracking control method. The dual-loop longitudinal tracking control method includes the control of the rail vehicle during autonomous driving based on speed information and the control of the rail vehicle during autonomous driving based on position information.

[0038] In one embodiment, after determining the reference speed curve, a dual-loop longitudinal tracking control method can be used to control the rail vehicle for autonomous driving. The dual-loop longitudinal tracking control includes two main control loops: a speed control loop and a position control loop. Among them, the speed control loop is based on the reference speed curve and the current actual speed information of the rail vehicle. The speed control loop is responsible for adjusting the acceleration or deceleration of the vehicle so that the actual speed of the vehicle gradually approaches and remains on the reference speed curve. This is usually achieved by controlling the traction and braking force of the vehicle. The position control loop is based on the current actual position information of the rail vehicle and the predetermined driving path. The position control loop is responsible for adjusting the direction and position of the vehicle to ensure that the vehicle can accurately travel along the predetermined path. This may require using the steering system of the vehicle and possibly the track switching mechanism to achieve. During the dual-loop longitudinal tracking control process, the speed control loop and the position control loop are coordinated and complementary to each other. The speed control loop is mainly responsible for maintaining the speed stability of the vehicle, while the position control loop is responsible for ensuring the accurate driving direction of the vehicle. Through the coordinated action of these two control loops, the rail vehicle can achieve autonomous driving with high precision and stability, realizing the efficient and safe autonomous driving control of the rail vehicle.

[0039] The present invention provides a method for autonomous driving of a rail vehicle applied to a rail vehicle. The rail vehicle travels on a vehicle road. The method includes: constructing the operation constraints for the rail vehicle to travel on the vehicle road; determining the reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraints; and controlling the rail vehicle for autonomous driving through a dual-loop longitudinal tracking control method based on the reference speed curve, achieving ensuring the minimum energy consumption of the rail vehicle and the optimal experience of passengers during the automatic driving process.

[0040] In another exemplary embodiment of the present invention, continuing with the example described above Figure 1 Taking the above-described embodiment as an example for illustration, among them, controlling the rail vehicle for autonomous driving through a dual-loop longitudinal tracking control method based on the reference speed curve can be achieved in the following manner: Based on the reference speed curve, determine the reference position curve corresponding to the reference speed curve; Based on the reference speed curve and the reference position curve, control the rail vehicle for autonomous driving through a dual-loop longitudinal tracking control method so that the deviation generated during the driving process of the rail vehicle is less than or equal to the deviation threshold, where the deviation includes the deviation in the speed dimension and the deviation in the position dimension.

[0041] In one embodiment, after determining the reference speed curve, a reference position curve corresponding to the reference speed curve can be further determined based on the reference speed curve and the driving path of the rail vehicle. The reference position curve represents the ideal positions that the vehicle should reach at different time points. It comprehensively considers the driving speed of the vehicle, the road geometric conditions, and the predetermined driving path to ensure that the vehicle can accurately travel at the predetermined speed and positions.

[0042] Furthermore, after determining the reference speed curve and the reference position curve, a double-loop longitudinal tracking control method can be adopted to perform automatic driving control on the rail vehicle, so that the deviation generated during the driving process of the rail vehicle is less than or equal to the deviation threshold, where the deviation includes the deviation in the speed dimension and the deviation in the position dimension. The deviation threshold can be adjusted according to the actual situation and is not specifically limited in this embodiment.

[0043] In the double-loop longitudinal tracking control process of this embodiment, the speed control loop and the position control loop are coordinated and complementary to each other. The speed control loop is mainly responsible for keeping the speed of the vehicle stable and close to the reference speed curve, while the position control loop is responsible for ensuring that the driving direction of the vehicle is accurate and close to the reference position curve. Through the synergistic effect of these two control loops and real-time monitoring and correction of deviations, the rail vehicle can achieve automatic driving with high precision and stability, while ensuring safety and efficiency during the driving process.

[0044] Figure 2 It is the second schematic flow chart of the rail vehicle automatic driving method provided by the present invention.

[0045] The following will be combined with Figure 2 to illustrate the process of another rail vehicle automatic driving method.

[0046] In an exemplary embodiment of the present invention, in combination with Figure 2 it can be known that after performing automatic driving control on the rail vehicle through the double-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the rail vehicle automatic driving method can further include steps 210 to 240, and each step will be introduced separately below.

[0047] In step 210, obtain the instantaneous speed of the rail vehicle at the current moment, and / or the instantaneous position of the rail vehicle at the current moment.

[0048] In step 220, determine the deviation of the rail vehicle at the current moment based on the instantaneous speed and the reference speed curve, and / or based on the instantaneous position and the reference position curve.

[0049] In one embodiment, during the automatic driving control process, the instantaneous speed and / or the instantaneous position of the rail vehicle at the current moment can be obtained in real time. Then, based on the instantaneous speed and the reference speed curve, and / or based on the instantaneous position and the reference position curve, the deviation of the rail vehicle at the current moment is determined. If the deviation is greater than a preset deviation threshold, the process proceeds to the next dynamic adjustment process.

[0050] In step 230, when the deviation is greater than the deviation threshold, the instantaneous operation information of the rail vehicle and the instantaneous road information of the vehicle road are obtained in real time, where the instantaneous operation information at least includes instantaneous acceleration information and instantaneous speed information; the instantaneous road information at least includes traffic light phase information and instantaneous speed limit information.

[0051] In step 240, based on the instantaneous operation information and the instantaneous road information, the reference speed curve is re-optimized to obtain an optimized reference speed curve, and the automatic driving control of the rail vehicle is performed based on the optimized reference speed curve.

[0052] In one embodiment, when it is detected that the deviation exceeds the deviation threshold, the instantaneous operation information of the rail vehicle and the instantaneous road information of the vehicle road can be obtained in real time. The instantaneous operation information at least includes instantaneous acceleration information and instantaneous speed information, which reflect the current dynamic state of the vehicle. The instantaneous road information at least includes traffic light phase information and instantaneous speed limit information, which reflect the current road environment state of the vehicle.

[0053] Furthermore, based on the instantaneous operation information and the instantaneous road information, the reference speed curve is re-optimized. The re-optimization process may involve adjusting the shape, smoothness, or speed value of the speed curve to adapt to the current road conditions, traffic conditions, and the dynamic characteristics of the vehicle. Through the re-optimization process, an optimized reference speed curve is obtained. During application, based on the optimized reference speed curve (and possible corresponding adjustments to the reference position curve), the automatic driving control of the rail vehicle can be performed, thereby ensuring that the rail vehicle can more accurately adapt to the current road environment and traffic conditions, and improving the accuracy and safety of automatic driving.

[0054] In yet another embodiment, based on the reference speed curve, a finite state machine model can be adopted for online speed planning. The state machine originates from the full operating conditions of the train operation, including the outbound departure state, the main line operation state, the vehicle following state, the intersection operation state, the inbound parking state, the platform departure state, the return depot parking state, and the abnormal scenario state. Among them, in the scenario of the outbound departure state machine, the train self-check result can be received and the speed planning algorithm can be self-started, and the road traffic flow speed and traffic light phase information can be obtained according to the V2X technology; then, the dynamic programming algorithm is used to plan the reference speed curve (corresponding to the reference speed curve) according to the high-precision map, V2X information, and the offline timetable. Further, based on the reference speed curve, according to the train positioning information, the expected speed to be adopted currently is calculated, and a PI controller with feed-forward compensation is used to achieve speed tracking.

[0055] In yet another embodiment, in the scenario of the main line operation state machine, the fast response and high-precision tracking of the reference speed curve can be achieved according to the current train positioning information; then, the time deviation of the train at the current position is calculated in real time (this deviation may be caused by the accumulated control deviation or may be caused by an abnormal scenario). When the deviation exceeds a certain threshold, dynamic programming will be triggered to regenerate the reference speed curve (corresponding to the optimized reference speed curve); further, the rail vehicle can be controlled for autonomous driving based on the optimized reference speed curve.

[0056] Figure 3 It is the third schematic flow chart of the rail vehicle autonomous driving method provided by the present invention.

[0057] To further introduce the rail vehicle autonomous driving method provided by the present invention, the following will be combined with Figure 3 for description.

[0058] In an exemplary embodiment of the present invention, in combination with Figure 3 it can be seen that after the rail vehicle is controlled for autonomous driving by means of a double-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the rail vehicle autonomous driving method may further include steps 310 to 340. The following will introduce each step separately: In step 310, when it is detected that there is an obstacle in front of the rail vehicle, the first running speed of the rail vehicle and the second running speed of the obstacle are obtained.

[0059] In step 320, based on the first running speed and the second running speed, the running acceleration of the rail vehicle is determined.

[0060] In one embodiment, during autonomous driving, sensors (such as radar, cameras, etc.) can be used to detect in real time whether there are obstacles in front of the rail vehicle. If an obstacle is detected, the current first running speed (instantaneous speed) of the rail vehicle and the second running speed of the front obstacle (when the obstacle is a moving object) can be obtained. Based on these two speed values, the running acceleration required for the rail vehicle to avoid collision or safely bypass can be calculated.

[0061] In step 330, based on the running acceleration, a temporary reference speed curve corresponding to the running acceleration is obtained.

[0062] In step 340, based on the temporary reference speed curve, autonomous driving control is performed on the rail vehicle.

[0063] In another embodiment, a temporary reference speed curve matching the current acceleration can be dynamically generated according to the determined running acceleration. This curve is designed to guide the rail vehicle to adjust its speed in a safe and stable manner to avoid collision with obstacles.

[0064] Furthermore, based on the generated temporary reference speed curve (and possibly the correspondingly adjusted reference position curve), autonomous driving control is performed on the rail vehicle. After the obstacle is safely bypassed or eliminated, the original reference speed curve and reference position curve are switched back for autonomous driving according to the actual situation. This embodiment significantly improves the safety and flexibility of the rail vehicle during autonomous driving by adding a detection and response mechanism for the front obstacle and dynamically generating a temporary reference speed curve for autonomous driving control, ensuring that the rail vehicle can safely and effectively respond to emergencies during autonomous driving.

[0065] In another exemplary embodiment of the present invention, continuing with the embodiment described above, after performing autonomous driving control on the rail vehicle based on the temporary reference speed curve, the rail vehicle autonomous driving method further includes the following steps: In the case where it is detected that the obstacle has left, autonomous driving control is performed on the rail vehicle by means of a double-loop longitudinal tracking control method based on the reference speed curve.

[0066] In one embodiment, once it is detected that the obstacle has left, autonomous driving control can be performed on the rail vehicle by means of a double-loop longitudinal tracking control method based on the previously determined reference speed curve and reference position curve. This step is aimed at enabling the rail vehicle to smoothly recover from the state of dealing with the obstacle to the normal autonomous driving state. This embodiment further improves the safety and stability of the rail vehicle during autonomous driving by adding a detection and processing mechanism for the departure of the obstacle and smoothly restoring the double-loop longitudinal tracking control based on the detection result.

[0067] In yet another embodiment, taking online speed planning using a finite state machine model as an example for illustration, in the scenario of a vehicle following state machine, when it is sensed that a dynamic obstacle enters the virtual track, the position and speed of the dynamic obstacle given by the sensing module can be obtained in real time; according to the state of the host vehicle and the state of the dynamic obstacle, the IntelligentDriver Model (IDM) model is used to calculate the train acceleration in real time, and a reference speed curve within the next 5 seconds (corresponding to the temporary reference speed curve) is generated in real time according to the train acceleration. After the dynamic obstacle leaves the virtual track, it can switch back to the main line operation state machine.

[0068] In yet another embodiment, in the scenario of an intersection operation state machine, based on the results of the high-precision map and train positioning, it can be determined whether to enter the intersection for operation. Before entering, the train will switch to the intersection state machine, and in this state machine, a PI controller with feedforward compensation will still be used; due to the complex intersection traffic environment, interaction scenarios with other vehicles may occur. Therefore, when the train is in the intersection state, a real-time discriminator needs to be added to determine whether the train needs to enter the following state and decelerate to follow; then, according to the train positioning, when leaving the intersection, it will switch to the main line operation state machine. In this state machine, it will be determined whether the reference speed curve is available. If not, dynamic programming will be triggered to plan a new reference speed curve. It can be understood that planning a new reference speed curve can be achieved by referring to the method of optimizing the reference speed curve described above.

[0069] In yet another embodiment, in the scenario of an in-station parking state machine, since in-station parking requires high precision and needs to meet a 99.995% probability that the station parking accuracy is within ±15 cm, a double-loop position controller will be used to improve the accuracy. The train will enter the in-station parking state machine at a certain distance in advance; a double-loop high-precision position controller can be designed. Among them, the inner loop is the speed loop, and the outer loop is the position loop. The outer loop generates a compensation speed through position deviation compensation, and the output control quantity is the train acceleration.

[0070] In yet another embodiment, in the scenario of a platform departure state machine, since it is necessary to confirm that the train doors are closed before departure at the platform, it is necessary to receive the train self-check signal to start the program automatically; since the number of passengers boarding at different stations is different and the time required for passengers to enter the train cannot be determined, in order to meet the on-time rate to the next station, each time the train departs from the platform, an algorithm for generating a reference speed curve based on dynamic programming needs to be triggered. After that, it enters the main line operation state machine.

[0071] In another embodiment, for the scenario of the return-to-depot parking state machine, similar to the inbound parking mode, a dual-loop high-precision position controller is adopted to achieve the purpose of precise parking. The parking accuracy is within ±15 cm with a 99.995% probability; it is necessary to consider the abnormal handling of possible positioning signal occlusion scenarios; it is necessary to judge whether the parking is completed according to the train's position and attitude, and send corresponding instructions after the parking task is completed.

[0072] In another embodiment, for the abnormal scenario state machine scenario, the abnormal scenario state machine may include communication abnormalities, etc. When a communication abnormality is detected (such as not receiving the heartbeat signal of the sensor within a certain period of time), it will enter the emergency braking state machine. At this time, the train will combine the current speed, heading angle, and yaw rate, and cooperate with the full-axis controller to give the train deceleration and steering wheel angle according to the road boundary obtained from the high-precision map.

[0073] Due to the different requirements of the scenarios, a scenario-adaptive longitudinal controller is proposed. Among them, in the scenarios of departure from the depot, running on the main line, vehicle following, passing through a green light at an intersection, and departure from the platform, a PI controller with feedforward compensation can be adopted; in the scenarios of inbound parking and return-to-depot parking, a dual-loop high-precision position controller can be adopted. In this embodiment, the control method is not specifically limited.

[0074] Figure 4 It is a schematic flowchart of the process for determining the reference speed curve of the rail vehicle running on the vehicle road based on the operation constraints provided by the present invention.

[0075] Next, in combination with Figure 4 The process of determining the reference speed curve of the rail vehicle running on the vehicle road based on the operation constraints will be described.

[0076] In an exemplary embodiment of the present invention, in combination with Figure 4 It can be known that determining the reference speed curve of the rail vehicle running on the vehicle road based on the operation constraints may include steps 410 to 450, and each step will be introduced separately below.

[0077] In step 410, determine the arrival operation time information of the rail vehicle in the road sub-interval of the vehicle road, and the speed limit information of the rail vehicle in the road sub-interval.

[0078] In one embodiment, the arrival operation time information of the rail vehicle in each road sub-interval can be determined according to the operation plan and road conditions of the rail vehicle, which includes the expected arrival time and residence time, etc. In the application process, the arrival operation time information can be obtained according to the historical operation information of the rail vehicle.

[0079] In one example, historical driving data can be obtained, mathematical methods can be used to analyze the external factors affecting the train operation speed, and a train operation scenario model can be established. First, organize and analyze the train operation scenario data, including road structure information (such as road gradient, traffic lights, lane speed limits, etc.) and virtual rail transit information (such as types of other traffic participants, time periods, frequencies, average speeds, etc.); second, based on the feature analysis of historical data, iteratively obtain the commuting time for the train to reach each station, including information such as commuting time during traffic peaks and commuting time during traffic valleys; finally, establish an offline train operation schedule, where the offline train operation schedule can be considered as the arrival operation time information.

[0080] In another example, the speed limit information for the rail vehicle in each road sub - interval can also be determined according to the characteristics of the road sub - interval (such as curves, slopes, bridges, etc.) and traffic rules.

[0081] In step 420, based on the arrival operation time information and the speed limit information, operation constraints are constructed.

[0082] In one embodiment, the arrival operation time information and the speed limit information can be used as operation constraints for the subsequent speed curve optimization process. During the application process, based on the actual operation data, external factors can be analyzed, including statistical information such as road gradient, traffic light position and time phase, types of other traffic participants occupying the road, time periods, durations, frequencies, speeds, etc., and external constraint conditions in the train operation curve planning process can be analyzed and established, including acceleration constraints, speed limit requirements, and punctuality constraints.

[0083] In step 430, a passenger comfort index function, an energy consumption index function, and a punctuality index function are respectively determined, and based on the passenger comfort index function, the energy consumption index function, and the punctuality index function, an objective optimization function under the road sub - interval is constructed, where the passenger comfort index function is determined according to the acceleration of the rail vehicle; the energy consumption index function is determined according to the traction force and / or braking force of the rail vehicle; and the punctuality index function is determined according to the arrival operation time information of the rail vehicle in the road sub - interval.

[0084] In one embodiment, a passenger comfort index function can be constructed according to the acceleration change of the rail vehicle. Generally, the smaller the acceleration change, the higher the passenger comfort. An energy consumption index function can be constructed according to the traction force and / or braking force of the rail vehicle. The magnitudes of the traction force and the braking force directly affect the energy consumption of the vehicle. A punctuality index function can also be constructed according to the arrival operation time information of the rail vehicle in each road sub - interval. The smaller the deviation between the actual arrival time and the expected arrival time of the vehicle, the higher the punctuality.

[0085] Further, the above three index functions can be weighted and summed to construct an objective optimization function. The selection of weights can be adjusted according to actual requirements to balance the relationship among passenger comfort, energy consumption, and punctuality.

[0086] In step 440, based on the objective optimization function and operation constraints, a sub-interval reference speed curve of the rail vehicle in the road sub-interval is determined to maximize the function value of the objective optimization function under the sub-interval reference speed curve.

[0087] In step 450, based on the sub-interval reference speed curves of the rail vehicle in each road sub-interval, a reference speed curve of the rail vehicle traveling on the vehicle road is determined.

[0088] In one embodiment, within each road sub-interval, based on the objective optimization function and operation constraints (including arrival operation time information and speed limit information), an optimization algorithm can be used to optimize the speed curve. The objective of the optimization is to maximize the function value of the objective optimization function, that is, to find a speed curve that not only satisfies the operation constraints but also achieves the best balance among passenger comfort, energy consumption, and punctuality. Then, the optimized speed curve is used as the sub-interval reference speed curve of this road sub-interval. Further, based on the sub-interval reference speed curves of the rail vehicle in each road sub-interval, a reference speed curve of the rail vehicle traveling on the entire vehicle road is obtained.

[0089] In this embodiment, by constructing operation constraints, constructing an objective optimization function, and determining a sub-interval reference speed curve, the determination of the reference speed curve of the rail vehicle traveling on the vehicle road is realized. This method comprehensively considers multiple aspects such as passenger comfort, energy consumption, and punctuality, and provides an effective speed control strategy for the automatic driving of rail vehicles.

[0090] The process of determining the sub-interval reference speed curve will be described below.

[0091] In another exemplary embodiment of the present invention, Figure 5 is a schematic diagram of the sub-interval reference speed curve provided by the present invention. Combining Figure 5 it can be seen that the curve trend of the sub-interval reference speed curve can successively be running at a constant speed according to the starting speed, accelerating from the starting speed to the maximum operable speed based on the first duration according to the target acceleration, maintaining the maximum operable speed and running at a constant speed, and decelerating from the maximum operable speed to the termination speed, where the starting speed is the running speed of the rail vehicle at the starting position of the road sub-interval; the termination speed is the running speed of the rail vehicle at the termination position of the road sub-interval.

[0092] Among them, the determination of the sub-interval reference speed curve can be achieved in the following manner: Determine the target acceleration and the first duration; Determine the sub-interval reference speed curve according to the curve trend of the sub-interval reference speed curve, in combination with the target acceleration and the first duration.

[0093] In one embodiment, for each road sub-interval, that is, the speed curve generation between two adjacent intersections / speed limit intervals, it can be solved based on the kinematic method. Assume that the speed limit at intersection 1 is v1 (the running speed corresponding to the starting position of the road sub-interval, that is, the starting speed), the speed limit at intersection 2 is v2 (the running speed corresponding to the ending position of the road sub-interval, that is, the ending speed), the length of intersection 1 is s1, and the distance from intersection 1 to intersection 2 after passing through intersection 1 is s2. As Figure 5 shown, the reference speed curve between two nodes can be divided into four stages: (1) The train arrives at intersection 1 at a speed of v1 at time t1 and passes through intersection 1 at a constant speed of v1; (2) Accelerate to vm (corresponding to the maximum operable speed) with an acceleration of a1 (corresponding to the target acceleration); (3) Keep running at a constant speed of vm for a period of time; (4) Decelerate to v2 with an acceleration of a2 and arrive at intersection 2.

[0094] Furthermore, the following kinematic equations (1) can be listed according to the above four stages: (1) It should be noted that ∆t1 in formula (1) corresponds to Figure 5 the duration between point a and point b in; ∆t2 corresponds to Figure 5 the duration between point b and point c in (corresponding to the first duration); ∆t3 corresponds to Figure 5 the duration between point c and point d in; ∆t4 corresponds to Figure 5 the duration between point d and point e in.

[0095] Among them, there are 7 unknowns and 5 equations in this system of equations, so the degree of freedom is 2. Select vm and ∆t2 as free variables, and the determined values can uniquely determine Figure 3 the speed curve in. That is, determine the target acceleration and the first duration; determine the sub-interval reference speed curve according to the curve trend of the sub-interval reference speed curve, in combination with the target acceleration and the first duration.

[0096] In another exemplary embodiment of the present invention, the target acceleration and the first duration can be achieved in the following manner: Sample multiple groups of candidate target accelerations and candidate first durations at a sampling interval, and based on the multiple groups of candidate target accelerations and candidate first durations, obtain multiple candidate sub-interval reference velocity curves corresponding to the multiple groups of candidate target accelerations and candidate first durations, and multiple candidate transfer costs corresponding to the multiple candidate sub-interval reference velocity curves; Based on the multiple candidate transfer costs, obtain the minimum candidate transfer cost; and based on the candidate sub-interval reference velocity curve corresponding to the minimum candidate transfer cost, determine the target acceleration and the first duration.

[0097] In one embodiment, considering comfort and energy consumption, the cost corresponding to the velocity curve can be calculated 。Sampling a series of values of vm and ∆t2 at a certain sampling interval can obtain a corresponding cluster of reference velocity curves and corresponding transfer costs, and then the minimum transfer cost can be found and its corresponding reference velocity curve. That is, among the multiple candidate transfer costs, obtain the minimum candidate transfer cost; and based on the candidate sub-interval reference velocity curve corresponding to the minimum candidate transfer cost, determine the target acceleration and the first duration.

[0098] As can be seen from the foregoing description, the present invention provides an automatic driving method for a rail vehicle applied to a rail vehicle traveling on a vehicle road. The method includes: constructing an operation constraint for the rail vehicle traveling on the vehicle road; determining a reference velocity curve for the rail vehicle traveling on the vehicle road based on the operation constraint; and performing automatic driving control on the rail vehicle through a double-loop longitudinal tracking control method based on the reference velocity curve, achieving ensuring the minimum energy consumption of the rail vehicle and the optimal experience of passengers during the automatic driving process.

[0099] The following describes the automatic driving device for a rail vehicle provided by the present invention. The automatic driving device for a rail vehicle described below can be correspondingly referred to the automatic driving method for a rail vehicle described above.

[0100] Figure 6 is a schematic structural diagram of the automatic driving device for a rail vehicle provided by the present invention.

[0101] In an exemplary embodiment of the present invention, the automatic driving device for a rail vehicle can be applied to a rail vehicle traveling on a vehicle road. Combining Figure 6 it can be known that the automatic driving device for a rail vehicle can include a construction module 610, a determination module 620, and a control module 630. Each module will be introduced separately below.

[0102] The construction module 610 can be configured to construct the operation constraint for the rail vehicle traveling on the vehicle road; A determination module 620, which can be configured to determine a reference speed curve for the rail vehicle to travel on the vehicle road based on the operating constraints; A control module 630, which can be configured to perform an autonomous driving control on the rail vehicle by a dual-loop longitudinal tracking control method based on the reference speed curve, wherein the dual-loop longitudinal tracking control method includes controlling the rail vehicle during the autonomous driving process based on speed information and controlling the rail vehicle during the autonomous driving process based on position information.

[0103] In an exemplary embodiment of the present invention, the control module 630 can implement the autonomous driving control on the rail vehicle by the dual-loop longitudinal tracking control method based on the reference speed curve in the following manner: Based on the reference speed curve, determine a reference position curve corresponding to the reference speed curve; Based on the reference speed curve and the reference position curve, perform an autonomous driving control on the rail vehicle by a dual-loop longitudinal tracking control method, so that the deviation generated during the driving process of the rail vehicle is less than or equal to a deviation threshold, wherein the deviation includes a deviation in the speed dimension and a deviation in the position dimension.

[0104] In an exemplary embodiment of the present invention, the control module 630 can also be configured to: Obtain the instantaneous speed of the rail vehicle at the current moment, and / or the instantaneous position of the rail vehicle at the current moment; Based on the instantaneous speed and the reference speed curve, and / or based on the instantaneous position and the reference position curve, determine the deviation of the rail vehicle at the current moment; In the case where the deviation is greater than the deviation threshold, obtain the instantaneous operation information of the rail vehicle and the instantaneous road information of the vehicle road in real time, wherein the instantaneous operation information at least includes instantaneous acceleration information and instantaneous speed information; the instantaneous road information at least includes the phase information of the traffic light and the instantaneous speed limit information; Based on the instantaneous operation information and the instantaneous road information, perform a re-optimization process on the reference speed curve to obtain an optimized reference speed curve, and perform an autonomous driving control on the rail vehicle based on the optimized reference speed curve.

[0105] In an exemplary embodiment of the present invention, the control module 630 can also be configured to: In the case where an obstacle is detected in front of the rail vehicle, obtain the first running speed of the rail vehicle and the second running speed of the obstacle; Determine the running acceleration of the rail vehicle based on the first running speed and the second running speed; Obtain a temporary reference speed curve corresponding to the running acceleration based on the running acceleration; Perform automatic driving control on the rail vehicle based on the temporary reference speed curve.

[0106] In an exemplary embodiment of the present invention, the control module 630 may further be configured to: When it is detected that the obstacle has left, perform automatic driving control on the rail vehicle based on the reference speed curve by means of double-loop longitudinal tracking control.

[0107] In an exemplary embodiment of the present invention, the construction module 610 may implement the construction of the running constraint in the following manner: Determine the arrival running time information of the rail vehicle in the road sub-interval of the vehicle road, and the speed limit information of the rail vehicle in the road sub-interval; Construct the running constraint based on the arrival running time information and the speed limit information; The determination module 620 may implement the determination of the reference speed curve for the rail vehicle to travel on the vehicle road based on the running constraint in the following manner: Respectively determine a passenger comfort index function, an energy consumption index function, and a punctuality index function, and construct an objective optimization function under the road sub-interval based on the passenger comfort index function, the energy consumption index function, and the punctuality index function, wherein the passenger comfort index function is determined according to the acceleration of the rail vehicle; the energy consumption index function is determined according to the traction force and / or braking force of the rail vehicle; the punctuality index function is determined according to the arrival running time information of the rail vehicle in the road sub-interval; Based on the objective optimization function and the running constraint, determine the sub-interval reference speed curve of the rail vehicle in the road sub-interval so that the function value of the objective optimization function under the sub-interval reference speed curve is the largest; Based on the sub-interval reference speed curves of the rail vehicle in each road sub-interval, determine the reference speed curve for the rail vehicle to travel on the vehicle road.

[0108] In an exemplary embodiment of the present invention, the curve trend of the sub-interval reference speed curve is successively running at a constant speed according to the starting speed, accelerating from the starting speed to the maximum operable speed based on a first duration according to the target acceleration, running at a constant speed while maintaining the maximum operable speed, and decelerating from the maximum operable speed to the termination speed, where the starting speed is the running speed of the rail vehicle at the starting position of the road sub-interval; the termination speed is the running speed of the rail vehicle at the termination position of the road sub-interval; The determination module 620 may implement the determination of the sub-interval reference speed curve in the following manner: Determine the target acceleration and the first duration; According to the curve trend of the sub-interval reference speed curve, in combination with the target acceleration and the first duration, determine the sub-interval reference speed curve.

[0109] In an exemplary embodiment of the present invention, the determination module 620 may implement the determination of the target acceleration and the first duration in the following manner: Sample multiple groups of candidate target accelerations and candidate first durations at a sampling interval, and based on the multiple groups of candidate target accelerations and candidate first durations, obtain multiple candidate sub-interval reference speed curves corresponding to the multiple groups of candidate target accelerations and candidate first durations, and multiple candidate transfer costs corresponding to the multiple candidate sub-interval reference speed curves; Based on the multiple candidate transfer costs, obtain the minimum candidate transfer cost; and based on the candidate sub-interval reference speed curve corresponding to the minimum candidate transfer cost, determine the target acceleration and the first duration.

[0110] Based on the same inventive concept, the present invention also provides a rail vehicle, which will be introduced below in combination with the following embodiments.

[0111] In an exemplary embodiment of the present invention, the rail vehicle may include a rail vehicle body and a processor, where the processor is configured to execute the rail vehicle automatic driving method described in any of the foregoing embodiments. Through this embodiment, it is realized to ensure the minimum energy consumption of the rail vehicle and the optimal experience of passengers during the automatic driving process.

[0112] Figure 7 Illustrates a schematic physical structure diagram of an electronic device, such as Figure 7As shown in the figure, the electronic device may include: a processor 710, a communications interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communications interface 720, and the memory 730 complete communication with each other through the communication bus 740. The processor 710 may call logic instructions in the memory 730 to execute an automatic driving method for a rail vehicle. The method is applied to a rail vehicle traveling on a vehicle road. The method includes: constructing an operation constraint for the rail vehicle to travel on the vehicle road; determining a reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraint; and performing automatic driving control on the rail vehicle through a double-loop longitudinal tracking control method. Among them, the double-loop longitudinal tracking control method includes controlling the rail vehicle during the automatic driving process based on speed information and controlling the rail vehicle during the automatic driving process based on position information.

[0113] In addition, when the logic instructions in the above-mentioned memory 730 are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0114] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the automatic driving method for a rail vehicle provided by each of the above methods. The method is applied to a rail vehicle that travels on a vehicle road, and the method includes: constructing an operation constraint for the rail vehicle to travel on the vehicle road; determining a reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraint; and performing automatic driving control on the rail vehicle through a dual-loop longitudinal tracking control method, where the dual-loop longitudinal tracking control method includes controlling the rail vehicle during the automatic driving process based on speed information and controlling the rail vehicle during the automatic driving process based on position information.

[0115] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is configured to execute the automatic driving method for a rail vehicle provided by each of the above methods. The method is applied to a rail vehicle that travels on a vehicle road, and the method includes: constructing an operation constraint for the rail vehicle to travel on the vehicle road; determining a reference speed curve for the rail vehicle to travel on the vehicle road based on the operation constraint; and performing automatic driving control on the rail vehicle through a dual-loop longitudinal tracking control method, where the dual-loop longitudinal tracking control method includes controlling the rail vehicle during the automatic driving process based on speed information and controlling the rail vehicle during the automatic driving process based on position information.

[0116] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0117] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatic driving of a rail vehicle, characterized in that: The method is applied to a rail vehicle, the rail vehicle traveling on a vehicle road, and the method comprises: Constructing operation constraints for the rail vehicle to travel on the vehicle road; determining a reference speed profile for the rail vehicle traveling on the vehicle road based on the operating constraints; Based on the reference speed curve, the rail vehicle is automatically driven through a dual-loop longitudinal tracking control method, wherein the dual-loop longitudinal tracking control method includes controlling the rail vehicle during the automatic driving process based on speed information and controlling the rail vehicle during the automatic driving process based on position information.

2. The rail vehicle automatic driving method according to claim 1, characterized in that: The automatic driving control of the rail vehicle by a dual-loop longitudinal tracking control method based on the reference speed curve specifically includes: Based on the reference speed curve, determining a reference position curve corresponding to the reference speed curve; Based on the reference speed curve and the reference position curve, the rail vehicle is automatically driven and controlled through a dual-loop longitudinal tracking control method so that the deviation generated by the rail vehicle during driving is less than or equal to a deviation threshold, wherein the deviation includes a deviation in the speed dimension and a deviation in the position dimension.

3. The rail vehicle automatic driving method according to claim 2, characterized in that: After the rail vehicle is automatically controlled by a dual-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the method further includes: Obtaining the instantaneous speed of the rail vehicle at the current moment, and / or the instantaneous position of the rail vehicle at the current moment; Determine the deviation of the rail vehicle at the current moment based on the instantaneous speed and the reference speed curve, and / or based on the instantaneous position and the reference position curve; In the case where the deviation is greater than the deviation threshold, real-time operation information of the rail vehicle and real-time road information of the vehicle road are acquired in real time, wherein the real-time operation information at least includes real-time acceleration information and real-time speed information; the real-time road information at least includes phase information of traffic lights and real-time speed limit information; Based on the real-time operation information and the real-time road information, the reference speed curve is re-optimized to obtain an optimized reference speed curve, and the rail vehicle is automatically driven and controlled based on the optimized reference speed curve.

4. The rail vehicle automatic driving method according to claim 2, characterized in that: After the rail vehicle is automatically controlled by a dual-loop longitudinal tracking control method based on the reference speed curve and the reference position curve, the method further includes: When an obstacle is detected in front of the rail vehicle, obtaining a first running speed of the rail vehicle and a second running speed of the obstacle; determining a running acceleration of the rail vehicle based on the first running speed and the second running speed; Based on the running acceleration, obtaining a temporary reference speed curve corresponding to the running acceleration; Based on the temporary reference speed curve, the rail vehicle is automatically controlled.

5. The rail vehicle automatic driving method according to claim 4, characterized in that: After the automatic driving control of the rail vehicle is performed based on the temporary reference speed curve, the method further includes: When it is detected that the obstacle has left, the rail vehicle is automatically controlled through a dual-loop longitudinal tracking control method based on the reference speed curve.

6. The rail vehicle automatic driving method according to any one of claims 1 to 5, characterized in that: The operational constraints are constructed in the following way: Determining the arrival time information of the rail vehicle in the road subsection of the vehicle road, and the speed limit information of the rail vehicle in the road subsection; constructing the operation constraint based on the arrival operation time information and the speed limit information; Determining a reference speed curve for the rail vehicle traveling on the vehicle road based on the operation constraint specifically includes: Determine a passenger comfort index function, an energy consumption index function, and a punctuality index function respectively, and construct a target optimization function under the road sub-interval based on the passenger comfort index function, the energy consumption index function, and the punctuality index function, wherein the passenger comfort index function is determined according to the acceleration of the rail vehicle; the energy consumption index function is determined according to the traction and / or braking force of the rail vehicle; the punctuality index function is determined according to the arrival time information of the rail vehicle in the road sub-interval; Based on the objective optimization function and the operation constraint, determining a sub-interval reference speed curve of the rail vehicle in the road sub-interval so as to maximize the function value of the objective optimization function under the sub-interval reference speed curve; Based on each sub-interval reference speed curve of the rail vehicle in each road sub-interval, a reference speed curve for the rail vehicle to travel on the vehicle road is determined.

7. The rail vehicle automatic driving method according to claim 6, characterized in that: The curve trends of the sub-interval reference speed curve are, in order, uniform operation according to the starting speed, accelerated operation from the starting speed to the maximum operable speed according to the target acceleration based on the first duration, uniform operation at the maximum operable speed, and decelerated operation from the maximum operable speed to the terminal speed, wherein the starting speed is the operating speed of the rail vehicle at the starting position of the road sub-interval; the terminal speed is the operating speed of the rail vehicle at the terminal position of the road sub-interval; The sub-interval reference speed curve is determined in the following manner: Determining the target acceleration and the first duration; The sub-interval reference speed curve is determined according to a curve trend of the sub-interval reference speed curve in combination with the target acceleration and the first duration.

8. The rail vehicle automatic driving method according to claim 7, characterized in that: The target acceleration and the first duration are determined in the following manner: Sampling a plurality of groups of candidate target accelerations and candidate first durations according to a sampling interval, and obtaining a plurality of candidate sub-interval reference speed curves corresponding to the plurality of groups of candidate target accelerations and candidate first durations, and a plurality of candidate transfer costs corresponding to the plurality of candidate sub-interval reference speed curves based on the plurality of groups of candidate target accelerations and candidate first durations; Based on the plurality of candidate transfer costs, obtaining a minimum candidate transfer cost; The target acceleration and the first duration are determined based on the candidate sub-interval reference speed curve corresponding to the minimum candidate transfer cost.

9. An automatic driving device for a rail vehicle, characterized in that: The device is applied to a rail vehicle, the rail vehicle travels on a vehicle road, and the device comprises: A construction module, used for constructing operation constraints of the rail vehicle traveling on the vehicle road; a determination module, configured to determine a reference speed profile for the rail vehicle traveling on the vehicle road based on the operation constraints; A control module is used to control the rail vehicle in an automatic driving process based on the reference speed curve through a dual-loop longitudinal tracking control method, wherein the dual-loop longitudinal tracking control method includes controlling the rail vehicle in an automatic driving process based on speed information and controlling the rail vehicle in an automatic driving process based on position information.

10. A rail vehicle, characterized in that: The rail vehicle comprises: The rail vehicle body, and A processor, wherein the processor is used to execute the rail vehicle automatic driving method as described in any one of claims 1 to 8.

Images