Method and ATO device for controlling a train using automated train operation

Abstract

The subject matter of the present invention is a computer-implemented method for controlling a train (1) traveling on a rail (4) towards a target position (5) using automated train operation (ATO), comprising the steps: calculating a first guidance curve (10) for automated train operation considering boundary conditions; controlling the driving of the train (1) according to the first guidance curve (10); determining an actual state of the train (1), a deviation (12) between the actual state and the first guidance curve (10), and any change in the boundary conditions; calculating a new guidance curve (15) when the deviation (12) exceeds a predetermined threshold or the change in the boundary conditions exceeds a predetermined threshold, wherein the new guidance curve (15) starts after an initial interval (16) at a future point (14); controlling the driving of the train (1) starting at the future point (14) according to the new guidance curve (15); and controlling the driving of the train (1) during the initial interval (16) according to an predetermined intermediate guidance curve (17). Furthermore, subject of the present invention is a corresponding automated train operation (ATO) device.

Description

Technical Field

[0001] The present invention relates to a computer-implemented method for controlling a train using automated train operation according to claim 1, an automated train operation (ATO) device according to claim 10, and a computer program product according to claim 15.Background Art

[0002] Train control systems are essential for ensuring safe and efficient operation of rail vehicles. These systems typically encompass various components and subsystems that work together to manage train movements, monitor track conditions, and enforce speed restrictions. One key aspect of modern train control is automated train operation (ATO), which aims to reduce human error and optimize train performance by automating certain driving functions.

[0003] ATO systems rely on sophisticated algorithms to calculate optimal speed profiles, known as guidance curves, for trains to follow. These guidance curves consider numerous factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The algorithms must continuously adapt to changing conditions and recalculate guidance curves as needed to maintain safe and efficient operation.

[0004] However, existing ATO systems face challenges in balancing responsiveness with computational efficiency. Recalculating guidance curves too frequently can strain onboard computing resources, while infrequent updates may lead to suboptimal performance. Additionally, there can be a delay between when a new guidance curve is calculated and when it can be implemented, potentially resulting in deviations from the intended trajectory.

[0005] Another issue is the transition between an existing guidance curve and a newly calculated one. Abrupt changes in speed or acceleration profiles can lead to passenger discomfort and increased wear on train components. Smoothing these transitions while still adhering to safety constraints and operational requirements presents a significant technical challenge.

[0006] It has been recognized that an improved method and device for controlling a train using automated train operation is needed that overcomes one or more of these problems.

[0007] EP 3 405 377 A1 discloses an ATO device for automated driving of a rail vehicle on tracks equipped with different track-side vehicle securing devices. The ATO device includes multiple ATP (Automatic Train Protection) interfaces for connecting to different on-board vehicle securing devices. This allows the ATO device to receive necessary information for automated driving from various ATP systems, enabling operation on different track infrastructures.

[0008] EP 1 466 803 B1 discloses a method and system for automatically controlling the running of a railway vehicle. The system calculates a speed profile based on various parameters including the vehicle's position, speed, and acceleration, as well as track data such as gradients and speed limits. The speed profile is continuously updated to account for changes in operating conditions. The system aims to optimize energy consumption while adhering to timetable constraints and ensuring passenger comfort. It employs a predictive control strategy, calculating optimal control actions over a future time horizon and applying the first control action of the optimal sequence. This approach allows for smooth transitions between different operating modes and helps maintain precise control over the vehicle's movement.

[0009] Based on the known computer-implemented methods for controlling a train using automated train operation, the present invention addresses the challenge of providing a computer-implemented method that enables more efficient recalculation of guidance curves while maintaining smooth train operation.

[0010] The invention solves this problem through a method according to claim 1.

[0011] The following paragraphs elaborate on definitions for the technical terms used in the claim set. The definitions should not be understood and limiting to the disclosure, but rather as hinting on some ways of understanding implementations of the invention while not excluding interpretations not mentioned in the definitions. Accordingly, other implementations are within the scope of the following paragraphs.

[0012] As used herein, the term "first guidance curve" may refer to an initial speed profile or trajectory calculated by an automated train operation (ATO) system for controlling the movement of a train. This curve may represent a planned sequence of states, like speeds or accelerations over distance or time, taking into account various factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The first guidance curve may serve as a reference for the train's operation until conditions warrant the calculation of a new guidance curve.

[0013] The term "train" as used in the claims refers to any rail vehicle capable of traveling on rails. This includes, but is not limited to, passenger trains, freight trains, locomotives, trams, metros, highspeed trains, and cargo trains. A train may consist of one or more connected rail vehicles, such as a locomotive pulling passenger cars or freight cars. In the context of automated train operation, the term "train" encompasses the entire rail vehicle unit that is being controlled as a single entity.

[0014] The term "rail" refers to the tracks on which the train travels. This includes standard railway tracks, as well as specialized tracks for trams, metros, or other rail-based transportation systems. The rail provides guidance and support for the train's wheels, allowing for controlled movement along a predetermined path.

[0015] "Automated train operation" (ATO) refers to a system or method for controlling the movement of a train with reduced or no human intervention. ATO systems are typically implemented as software modules installed on an on-board device within the train itself, ensuring real-time control and responsiveness to local conditions. However, these systems can also be implemented on a server or in a cloud environment. A server in this context refers to a computer system dedicated to running specific software applications and services, while a cloud environment refers to a network of remote servers hosted on the internet to store, manage, and process data. The on-board implementation is often preferred for its reliability and low-latency performance, particularly in areas with limited network connectivity. It's important to note that ATO components can also be implemented in hardware, such as dedicated computer systems with processors, memory, and data storage devices specifically designed for train control applications, whether located on-board the train or in centralized control centers.

[0016] A "guidance curve" in the context of ATO refers to a calculated trajectory or speed profile that the train should follow. This curve considers various factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The guidance curve is typically represented as a set of data points defining the desired speed of the train at various positions along its route.

[0017] The term "boundary conditions" encompasses various constraints and parameters that affect the calculation of the guidance curve. These may include speed limits, track gradients, curves, station locations, power limits, and scheduled arrival times. Changes in boundary conditions could involve updates to any of these parameters that would necessitate recalculation of the guidance curve.

[0018] "Deviation" refers to the difference between the train's actual state, e.g. position, time or speed, and the state prescribed by the current guidance curve. This deviation is continuously or at least periodically monitored to determine if recalculation of the guidance curve is necessary.

[0019] The "initial interval" is a period of time or distance immediately following the decision to recalculate the guidance curve, during which the train continues to follow a predetermined intermediate guidance curve. This interval allows time for the new guidance curve to be calculated without requiring an immediate change in the train's behavior.

[0020] The "intermediate guidance curve" is a temporary trajectory or speed profile that the train follows during the initial interval. This could be based on the existing first guidance curve, a fixed value (such as a constant speed or acceleration), or another predefined profile that doesn't need much computational effort.

[0021] In the context of the method claims, each step can be considered as a module in a software or hardware implementation. For example, the step of calculating a guidance curve could be performed by a calculation module, which transforms input data (such as boundary conditions and train characteristics) into output data (the guidance curve). This output data would then be sent as a digital data telegram to a control module by means of a communication module. The control module would receive this data through another communication module and use it to control the train's movement.

[0022] The "target position" refers to a specific location along the rail that the train is intended to reach. This could be a station stop, a point where a speed change is required, or any other significant position along the train's route. The ATO system aims to control the train's movement to arrive at this target position according to the calculated guidance curve.Summary of Invention

[0023] The invention as described in claim 1 provides several significant advantages over previous solutions for controlling trains using automated train operation (ATO): 1. Improved computational efficiency: By calculating a new guidance curve that starts after an initial interval at a future point, the method allows for more efficient use of onboard computing resources. This approach reduces the strain on the ATO system's processor, as it does not need to immediately implement a newly calculated curve. This is particularly advantageous for embedded systems with limited computational power. 2. Continuous operation during recalculation: The use of a predetermined intermediate guidance curve during the initial interval ensures that the train continues to operate smoothly while a new guidance curve is being calculated. This eliminates any potential gaps or interruptions in train control that might occur if the system had to wait for a new curve to be fully computed before continuing operation. 3. Adaptive responsiveness: The method allows for dynamic adjustment of the train's trajectory by recalculating the guidance curve when deviations exceed a predetermined threshold or when boundary conditions change significantly. This ensures that the ATO system can respond effectively to changing conditions while avoiding unnecessary recalculations for minor fluctuations. 4. Smooth transitions: By starting the new guidance curve at a future point after an initial interval, the method allows for smoother transitions between guidance curves, i.e. from the first guidance curve to the new guidance curve. This can lead to improved passenger comfort and reduced wear on train components by avoiding abrupt changes in speed or acceleration. The future point is set taking the predetermined intermediate guidance curve into account, ensuring a seamless transition from the intermediate curve to the new guidance curve. 5. Flexibility in handling boundary condition changes: The method explicitly accounts for changes in boundary conditions, allowing the ATO system to adapt to updates in track conditions, speed limits, or other operational parameters. This ensures that the train always operates according to the most current information available. 6. Balance between responsiveness and stability: The use of predetermined thresholds for deviation and boundary condition changes strikes a balance between being responsive to significant changes and maintaining stable operation. This helps prevent oscillations or frequent unnecessary recalculations that could result from an overly sensitive system. 7. Proactive control: By calculating a new guidance curve that starts at a future point, the method allows the ATO system to anticipate and prepare for upcoming changes in the train's trajectory. This proactive approach can lead to more optimal control and potentially improved energy efficiency. 8. Scalability: The method is applicable to various types of rail vehicles and track configurations, as it does not rely on specific hardware or infrastructure beyond basic ATO capabilities. This makes it a versatile solution that can be implemented across different rail systems. 9. Real-time adaptation: The continuous monitoring of the train's actual state, e.g. position or time, and its deviation from the current guidance curve allows for real-time adaptation to unexpected situations, such as slight delays or variations in train performance. 10. Reduced reliance on frequent communication: By enabling the onboard ATO system to recalculate guidance curves independently, the method reduces the need for frequent communication with centralized control systems. This can be particularly advantageous in areas with limited or unreliable communication infrastructure.

[0024] These advantages collectively contribute to a more robust, efficient, and adaptable ATO system that can improve the overall performance and reliability of automated train operations.Description of Embodiments

[0025] In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to the first guidance curve. This approach offers several advantages. By basing the intermediate guidance curve on the existing first guidance curve, the method ensures a smooth transition during the initial interval while the new guidance curve is being calculated. This continuity helps maintain passenger comfort and reduces unnecessary wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve as a basis for the intermediate curve leverages the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period of the initial interval. Alternative implementations could involve interpolating between key points of the first guidance curve to generate a simplified intermediate curve, or using a subset of the first guidance curve's data points to create a less computationally intensive version for the initial interval.

[0026] In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to a fixed value, for example fixed velocity or fixed acceleration. E.g. the fixed velocity could be the actual velocity at the time when the need for the new guidance curve is derived. This approach offers simplicity and predictability during the initial interval. By maintaining a constant speed or acceleration, the train's behavior becomes highly predictable, which can be advantageous for energy management and passenger comfort. This method also reduces computational requirements during the initial interval, as maintaining a fixed value is less demanding than following a complex curve. Alternative implementations could involve using different fixed values based on the current section of track (e.g., higher fixed speeds on straight sections, lower fixed speeds on curves), or gradually transitioning between fixed values to create a stepped intermediate guidance curve.

[0027] In a further preferred embodiment of the inventive computer-implemented method, the length of the initial interval is set according to at least one current state of the train or at least one boundary condition. This adaptive approach to determining the initial interval length offers increased flexibility and responsiveness to real-time conditions. By considering the train's current state (such as speed, acceleration, or position) or relevant boundary conditions (such as upcoming track geometry or speed restrictions), the method can optimize the length of the initial interval to balance computational needs with operational efficiency. For example, a longer initial interval might be set when approaching a complex track section that requires more extensive calculations for the new guidance curve. Alternative implementations could involve using machine learning algorithms to dynamically adjust the initial interval length based on historical performance data, or implementing a multi-factor scoring system that weighs various states and conditions to determine the optimal interval length.

[0028] In a further preferred embodiment of the inventive computer-implemented method, the length of the initial interval is set according to an estimated time for calculating the new guidance curve. This approach directly links the initial interval to the computational requirements of generating the new guidance curve, ensuring that the train continues to operate smoothly while calculations are in progress. By estimating the calculation time, the method can adapt to varying computational loads or complexities in different scenarios, potentially improving overall system responsiveness. This could be particularly advantageous when dealing with sections of track that require more complex calculations due to multiple boundary condition changes or intricate geometries. Alternative implementations might involve using a rolling average of recent calculation times to estimate future needs, or implementing a predictive model that considers factors like upcoming track complexity and current computational load to estimate calculation times more accurately.

[0029] In a further preferred embodiment of the inventive computer-implemented method, the threshold for the change in the boundary conditions is zero in order to calculate the new guidance curve for every change in the boundary conditions. This approach ensures maximum responsiveness to any alterations in the operational environment. By recalculating the guidance curve for every detected change, no matter how small, the method can provide the most up-to-date and optimized trajectory for the train at all times. This could be particularly beneficial in complex rail environments with frequently changing conditions, such as dense urban networks or areas with variable speed restrictions. However, this approach may increase computational load and could potentially lead to more frequent transitions between guidance curves. Alternative implementations could involve setting non-zero thresholds for different types of boundary conditions based on their relative impact on train operation, or implementing a cumulative change threshold that triggers recalculation only when the sum of multiple small changes exceeds a certain value.

[0030] In a further preferred embodiment of the inventive computer-implemented method, the threshold for the deviation is dynamically adjusted based on at least one of: current speed of the train, distance to the target position, or complexity of upcoming track sections. This adaptive threshold approach allows for more nuanced control over when recalculations are triggered. By considering factors like current speed, remaining distance, or track complexity, the method can balance the need for accuracy with computational efficiency. For example, tighter deviation thresholds could be set when approaching the target position or when navigating complex track sections, while allowing for larger deviations at high speeds on straight sections where minor variations have less impact. Alternative implementations could involve using a multi-dimensional threshold that considers multiple factors simultaneously, or implementing a predictive model that adjusts thresholds based on anticipated future conditions along the route.

[0031] In a further preferred embodiment of the inventive computer-implemented method, the predetermined threshold for the deviation is a function of the distance to the target position. This approach recognizes that the importance of precise adherence to the guidance curve typically increases as the train approaches its target position. By making the deviation threshold a function of remaining distance, the method can become increasingly stringent in its control as the train nears its destination. This can help ensure accurate stopping at platforms or precise positioning at critical points along the route. For example, the threshold could decrease linearly or exponentially as the target position approaches. Alternative implementations might involve using a step function with discrete threshold changes at key distances, or combining the distance-based function with other factors like current speed or track geometry to create a more complex, multi-variable threshold function.

[0032] In a further preferred embodiment of the inventive computer-implemented method, the method further comprises adjusting the speed of the train during the initial interval to facilitate a smooth transition to the new guidance curve. This additional step enhances the overall smoothness of the train's operation during guidance curve transitions. By proactively adjusting the train's speed during the initial interval, the method can minimize potential discontinuities when switching to the new guidance curve. This can lead to improved passenger comfort, reduced wear on train components, and potentially better energy efficiency by avoiding sudden changes in power demand. The speed adjustment could be implemented as a gradual acceleration or deceleration to better align with the anticipated starting point of the new guidance curve. Alternative implementations could involve using predictive algorithms to estimate the likely characteristics of the new guidance curve and adjusting speed accordingly, or implementing a multi-stage transition process with increasingly fine-tuned speed adjustments as the end of the initial interval approaches.

[0033] Based on known devices for train control, the invention further addresses the challenge of providing an arrangement that enables efficient and accurate calculation of guidance curves for autonomous rail vehicles while maintaining smooth operation and reducing computational load. The invention solves this problem through an ATO device according to claim 10. Preferred embodiments result from the dependent claims 11 to 14, whereby the same advantages as initially explained for the inventive computer-implemented method apply analogously.

[0034] Based on known computer program products for calculation of guidance curves, the invention further addresses the challenge of providing a computer program product that enables efficient and accurate calculation of guidance curves for autonomous rail vehicles while maintaining smooth operation and reducing computational load.

[0035] The invention solves this problem through a computer program product according to claim 15. The same advantages as initially explained for the inventive computer-implemented method apply analogously.

[0036] Non-limiting and non-exhaustive examples are described with reference to the following Figures.Brief Description of the Drawings

[0037] For better explanation of the invention, the following schematic representations show: Fig. 1an exemplary embodiment of a rail vehicle system according to the invention, Fig. 2a velocity-distance graph showing guidance curves according to an embodiment of the invention, Fig. 3another alternative velocity-distance graph showing alternative guidance curves according to an embodiment of the invention, and Fig. 4a flowchart of an automated train operation process according to an embodiment of the invention performed by the train in Figure 1. Description of Examples

[0038] Figure 1 illustrates an exemplary embodiment of a rail vehicle system according to the invention. The system comprises a train 1 consisting of a locomotive 2 and a passenger car 3 traveling on a rail 4. The train 1 is moving towards a target position 5 as indicated by a direction indicator 6.

[0039] The locomotive 2 is equipped with a train control system 9 comprising an ATP device 7 and an ATO device 8. The ATO device 8 is configured according to an exemplary embodiment of the invention. The train control system 9 is positioned inside of the locomotive 2. The passenger car 3 is coupled to the locomotive 2, representing a typical configuration for passenger rail transport. The rail 4 provides the track on which the train 1 operates, guiding its movement towards the target position 5.

[0040] The target position 5 is marked on the rail 4, indicating a specific point the train 1 is approaching. The target position 5 may represent a station stop, speed restriction point, or other significant location for the train's operation. The direction indicator 6 shows the train's direction of travel, which is towards the target position 5.

[0041] The ATO device 8 is configured to calculate a first guidance curve 10 for automated train operation taking boundary conditions into account. These boundary conditions may include track geometry, speed limits, and other operational parameters. The ATO device 8 controls the automated driving of the train 1 according to this first guidance curve 10. The calculation of guidance curves according to the invention is described later in more detail.

[0042] The ATO device 8 may comprise a computer on which the inventive computer-implemented method is performed, allowing for real-time processing and execution of the automated train operation algorithms. This computer may include processors, memory, and storage components necessary to carry out the complex calculations and decision-making processes involved in generating and adjusting guidance curves.

[0043] The train control system 9 is capable of deriving information about boundary conditions like actual information points on the route based on stored route information and position information about the current position of the train 1. This capability allows the system to anticipate and respond to upcoming track features or operational requirements.

[0044] To obtain accurate position information, the train control system 9 may have an interface to additional systems of the train 1. These additional systems may include for example an odometer system, a Balise Transmission Module (BTM), or a Global Navigation Satellite System (GNSS) device. By integrating data from these sources, the train control system 9 can maintain precise knowledge of the train's actual position 11 along the rail 4.

[0045] The ATP device 7 works in conjunction with the ATO device 8 to ensure safe operation of the train 1. While the ATO device 8 manages the automated driving functions, the ATP device 7 provides an additional layer of safety oversight, monitoring the train's speed and position to prevent unsafe conditions.

[0046] Figure 4 illustrates a flowchart of a method for controlling the train 1 using automated train operation (ATO). The method is performed by the ATO device 8 of the train 1 introduced in Figure 1. Some parts are shown in Figures 2 and 3, which will be described in more detail below.

[0047] The process begins with a method step 100 of calculating a first guidance curve 10 considering boundary conditions. These boundary conditions may include track geometry, speed limits, and other operational parameters relevant to the train 1 traveling on the rail 4 towards the target position 5.

[0048] The next method step 110 involves controlling train driving according to the first guidance curve 10. The ATO device 8 uses this first guidance curve 10 to manage the automated operation of the train 1.

[0049] The process then moves to a method step 120 of determining an actual state like a position 11 of the train 1, a deviation 12 between the actual state and the first guidance curve 10, and any change in the boundary conditions. This step allows the ATO device 8 to monitor the train's progress and identify any significant deviations from the planned route according to the first guidance curve 10 or any changes in operational conditions. As this step is done continuously or at least periodically, the actual position 11 is shown in Figures 2 and 3 as trajectories. The deviation 12 is shown in the examples in Figures 2 and 3 as a deviation between positions, i.e. the actual position 11 and the position in the first guidance curve 10. However, in alternative embodiments the deviation could also be in time or speed.

[0050] A decision point 160 follows, where the process checks if the deviation 12 or boundary condition change exceeds a predetermined threshold respectively. If no threshold is exceeded, the process loops back to method step 110, continuing to control the train 1 according to the first guidance curve 10.

[0051] If any of the thresholds is exceeded, the process proceeds to a method step 130 to calculate a new guidance curve 15. This new guidance curve 15 starts after an initial interval 16 at a future point 14, allowing time for the calculations to be completed while the train 1 continues to operate. The future point 14 can be precisely determined since the intermediate guidance curve 17 is known, allowing the ATO device 8 to calculate the new guidance curve 15 to start exactly at this predetermined future point 14.

[0052] The length of the initial interval 16 may be set dynamically based on various factors to optimize the train's operation and ensure smooth transitions between the first guidance curve 10 and new guidance curve 15. For example, the length of the initial interval 16 may be determined by considering the current operational state of the train 1, including its speed, position, and acceleration. For instance, a longer initial interval 16 may be set when the train 1 is traveling at higher speeds to allow more time for calculations and gradual adjustments.

[0053] In other implementations, the initial interval 16 may be influenced by the complexity of the upcoming track section. If the train 1 is approaching a section with multiple curves, gradient changes, or speed restrictions, the ATO device 8 may set a longer initial interval 16 to ensure sufficient time for calculating a more complex new guidance curve 15. Conversely, for simpler track sections, a shorter initial interval 16 may be adequate.

[0054] The ATO device 8 may also consider the computational resources available when setting the initial interval 16. If the device is handling multiple tasks simultaneously or operating in a resource-constrained environment, it may extend the initial interval 16 to ensure adequate time for calculations without compromising other critical functions.

[0055] In some cases, the initial interval 16 may be adjusted based on historical data and machine learning algorithms. The ATO device 8 may analyze past performance and optimize the length of the initial interval 16 for different scenarios, potentially improving efficiency over time as it learns from previous operations.

[0056] The system may also consider external factors when setting the initial interval 16. For example, if the train 1 is operating in an area with known communication challenges or intermittent data updates, a longer initial interval 16 may be set to accommodate potential delays in receiving updated boundary conditions or other critical information.

[0057] In certain implementations, the initial interval 16 may be set as a function of the magnitude of the deviation 12 or the extent of changes in boundary conditions. Larger deviations or more significant changes may warrant longer initial intervals to allow for more comprehensive recalculations and smoother transitions.

[0058] The ATO device 8 may also implement adaptive strategies for setting the initial interval 16. It may start with a default interval length and dynamically adjust it based on the success of previous transitions between guidance curves. If smooth transitions are consistently achieved with shorter intervals, the system may gradually reduce the interval length to improve responsiveness. Conversely, if transitions are causing discomfort or operational issues, the system may increase the interval length to allow for more gradual adjustments.

[0059] The next method step 140 involves controlling the train 1 during the initial interval 16 using a predetermined intermediate guidance curve 17. This ensures continuous and smooth operation of the train 1 while the new guidance curve 15 is being calculated and the train 1 reaches the future point 14.

[0060] The intermediate guidance curve 17 may be set using various approaches to ensure smooth operation during the initial interval 16. For example, the ATO device 8 may set the intermediate guidance curve 17 based on the first guidance curve 10, potentially using a simplified version or subset of data points from the first guidance curve 10. This approach may provide continuity in the train's operation while reducing computational load during the transition period. In other implementations, the intermediate guidance curve 17 may be set according to a fixed value, such as a constant velocity or acceleration. This method may be particularly useful in situations where maintaining a steady state is preferable, for example, when approaching a station or navigating a complex track section. The ATO device 8 may select an appropriate fixed value based on the train's current speed, upcoming track conditions, or energy efficiency considerations. The ATO device 8 may also employ more sophisticated methods to set the intermediate guidance curve 17. For instance, it may use a predictive algorithm that considers the train's current state, upcoming track geometry, and anticipated characteristics of the new guidance curve 15 to generate an optimized intermediate trajectory. Anyway, the intermediate guidance curve 17 links to the future point 18, so that the start of the new guidance curve 15 is known.

[0061] Following this, the process moves to a method step 150 of controlling the train 1 starting at the future point 14 using the new guidance curve 15. This allows the ATO device 8 to implement the updated new guidance curve 15 based on the most recent calculations and conditions.

[0062] After this step, the process loops back to method step 120 as described above, continuing the cycle of monitoring, evaluating, and adjusting the train's operation as needed. Note that, after looping back to step 120, the new guidance curve 15 will be handled as the first guidance curve 10 before.

[0063] The flowchart in Figure 4 highlights the iterative nature of the method, with continuous monitoring and adjustment of the guidance curve based on real-time conditions. The process incorporates an adaptive mechanism that allows for recalculation of the guidance curve when significant deviations or changes in boundary conditions occur, while maintaining control during the recalculation period using the intermediate guidance curve 17.

[0064] This method, as performed by the ATO device 8, works in conjunction with the ATP device 7 and other components of the train control system 9 to ensure safe and efficient operation of the train 1 as it travels along the rail 4 towards the target position 5.

[0065] A computer program product may comprise instructions that, when executed by a computer, cause the computer to perform this method for controlling the train 1 using automated train operation. This computer program product may be integrated into the ATO device 8 or other suitable components of the train control system 9.

[0066] Figures 2 and 3 illustrate exemplary velocity-distance graphs showing guidance curves calculated and used by the ATO device 8 of the train 1 introduced in Figure 1. These graphs represent alternative outcomes when applying the method shown in Figure 4.

[0067] Figure 2 depicts a scenario where the new guidance curve 15 is calculated when the deviation 12 exceeds the predetermined threshold. The graph shows the first guidance curve 10 representing the planned speed of the train 1 over distance. The actual position 11 curve shows the real speed of the train 1 over distance. At the first position 13, the deviation 12 between the actual position 11 and the first guidance curve 10 exceeds the predetermined threshold. Consequently, as described above the ATO device 8 calculates the new guidance curve 15.

[0068] The threshold for the deviation 12 may be dynamically adjusted based on factors such as the current speed of the train 1, the distance to the target position 5, or the complexity of upcoming track sections. This adaptive approach allows for more precise control in critical areas while allowing for greater flexibility in less critical sections. The predetermined threshold for the deviation 12 may be set as a function of the distance to the target position 5. This approach typically results in tighter control as the train 1 approaches its destination.

[0069] The new guidance curve 15 starts at the future point 14, which is in the future relative to the first position 13. The distance between the first position 13 and the future point 14 represents the initial interval 16. During this initial interval 16, the ATO device 8 calculates the new guidance curve 15 while controlling the train 1 using the intermediate guidance curve 17.

[0070] In the exemplary situation shown in Figure 2 the intermediate guidance curve 17 is set to a constant velocity, which is the velocity at the first position 13. Here, the future point 14 may be calculated with a high degree of accuracy, as the velocity during the initial interval 16 is fixed. This fixed velocity allows for a precise determination of the train's position at the end of the initial interval 16. The ATO device 8 may use this information to set the starting point for the new guidance curve 15, ensuring that it begins exactly at the future point 14. By calculating the new guidance curve 15 to start at this predetermined future point 14, the system can provide a seamless transition from the intermediate guidance curve 17 to the new guidance curve 15. This approach may allow for smoother train operation and more efficient use of computational resources, as the ATO device 8 can focus on calculating the optimal trajectory from the known future starting point, which is the future point 14.

[0071] Alternatively, the intermediate guidance curve 17 may be set to a constant acceleration if this is deemed appropriate for the current operational context. This approach may be particularly useful in situations where maintaining a constant velocity is not ideal, such as when the train 1 is on an incline or decline, or when energy efficiency considerations favor a gradual change in speed. By setting a constant acceleration, the ATO device 8 may be able to more smoothly transition between the first guidance curve 10 and the new guidance curve 15, potentially improving passenger comfort and reducing wear on the train's components.

[0072] The ATO device 8 may calculate the intermediate guidance curve 17 based on various factors and constraints. In some implementations, the device may analyze the current state of the train 1, including its speed, position, and acceleration, as well as upcoming track conditions and operational requirements. The ATO device 8 may then use this information to determine the most appropriate parameters for the intermediate guidance curve 17, whether it be a constant velocity, constant acceleration, or a more complex profile. The calculation may also consider the estimated duration of the initial interval 16, ensuring that the intermediate guidance curve 17 provides a suitable trajectory for the train 1 until the new guidance curve 15 can be implemented at the second position point 14. Another alternative approach in getting the intermediate guidance curve 17 is now described in relation to Figure 3.

[0073] Figure 3 illustrates an alternative scenario where the new guidance curve 15 is calculated due to a change in a boundary condition 18 exceeding a predetermined threshold. The graph shows the first guidance curve 10 starting from a starting position S0 and increasing in velocity. At a first position 13 at position S1, there is a deviation 12 between the actual position 11 and the first guidance curve 10. But the deviation 12 is not above the threshold. Therefore, this is not the reason for calculation a new guidance curve 15. But at the first position 13, one of the boundary conditions 18 changes, specifically the speed limit Vmax that changes from Vmax to Vmax'.

[0074] In response to this change, the ATO device 8 calculates the new guidance curve 15. The new guidance curve 15 begins at a future point 14 at position S2, after the initial interval 16 between S1 and S2. The new guidance curve 15 reaches a plateau at the new speed limit Vmax' and then decreases as the train 1 approaches the target position 5.

[0075] The length of the initial interval 16 may be set according to an estimated time for calculating the new guidance curve 15. This allows the ATO device 8 to complete the necessary calculations before implementing the new guidance curve 15.

[0076] In some implementations, the threshold for changes in the boundary conditions 18 may be set to zero. This approach ensures that the ATO device 8 calculates a new guidance curve 15 for every change in the boundary conditions 18, no matter how small.

[0077] The intermediate guidance curve 17 used during the initial interval 16 in the exemplary embodiment in Figure 3 is set by the ATO device 8 according to the first guidance curve 10. This approach may offer several advantages. By basing the intermediate guidance curve 17 on the existing first guidance curve 10, the method may ensure a smooth transition during the initial interval 16 while the new guidance curve 15 is being calculated. This continuity may help maintain passenger comfort and reduce wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve 10 as a basis for the intermediate curve 17 may leverage the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period. If necessary, the ATO device 8 may interpolate between key points of the first guidance curve 10 to generate a simplified intermediate curve 17, or use a subset of the first guidance curve's data points to create a less computationally intensive version for the initial interval 16.

[0078] The length of the initial interval 16 may be set according to at least one current state of the train 1 or at least one boundary condition 18. This allows the ATO device 8 to adapt the calculation time based on the current operational context.

Examples

Embodiment Construction

[0025]In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to the first guidance curve. This approach offers several advantages. By basing the intermediate guidance curve on the existing first guidance curve, the method ensures a smooth transition during the initial interval while the new guidance curve is being calculated. This continuity helps maintain passenger comfort and reduces unnecessary wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve as a basis for the intermediate curve leverages the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period of the initial interval. Alternative implementations could involve interpolating between key points of the first guidance curve to generate a simplified intermediate curve, or using a subset of the first guidance cur...

Technical Field

[0001] The present invention relates to a computer-implemented method for controlling a train using automated train operation according to claim 1, an automated train operation (ATO) device according to claim 10, and a computer program product according to claim 15.Background Art

[0002] Train control systems are essential for ensuring safe and efficient operation of rail vehicles. These systems typically encompass various components and subsystems that work together to manage train movements, monitor track conditions, and enforce speed restrictions. One key aspect of modern train control is automated train operation (ATO), which aims to reduce human error and optimize train performance by automating certain driving functions.

[0003] ATO systems rely on sophisticated algorithms to calculate optimal speed profiles, known as guidance curves, for trains to follow. These guidance curves consider numerous factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The algorithms must continuously adapt to changing conditions and recalculate guidance curves as needed to maintain safe and efficient operation.

[0004] However, existing ATO systems face challenges in balancing responsiveness with computational efficiency. Recalculating guidance curves too frequently can strain onboard computing resources, while infrequent updates may lead to suboptimal performance. Additionally, there can be a delay between when a new guidance curve is calculated and when it can be implemented, potentially resulting in deviations from the intended trajectory.

[0005] Another issue is the transition between an existing guidance curve and a newly calculated one. Abrupt changes in speed or acceleration profiles can lead to passenger discomfort and increased wear on train components. Smoothing these transitions while still adhering to safety constraints and operational requirements presents a significant technical challenge.

[0006] It has been recognized that an improved method and device for controlling a train using automated train operation is needed that overcomes one or more of these problems.

[0007] EP 3 405 377 A1 discloses an ATO device for automated driving of a rail vehicle on tracks equipped with different track-side vehicle securing devices. The ATO device includes multiple ATP (Automatic Train Protection) interfaces for connecting to different on-board vehicle securing devices. This allows the ATO device to receive necessary information for automated driving from various ATP systems, enabling operation on different track infrastructures.

[0008] EP 1 466 803 B1 discloses a method and system for automatically controlling the running of a railway vehicle. The system calculates a speed profile based on various parameters including the vehicle's position, speed, and acceleration, as well as track data such as gradients and speed limits. The speed profile is continuously updated to account for changes in operating conditions. The system aims to optimize energy consumption while adhering to timetable constraints and ensuring passenger comfort. It employs a predictive control strategy, calculating optimal control actions over a future time horizon and applying the first control action of the optimal sequence. This approach allows for smooth transitions between different operating modes and helps maintain precise control over the vehicle's movement.

[0009] Based on the known computer-implemented methods for controlling a train using automated train operation, the present invention addresses the challenge of providing a computer-implemented method that enables more efficient recalculation of guidance curves while maintaining smooth train operation.

[0010] The invention solves this problem through a method according to claim 1.

[0011] The following paragraphs elaborate on definitions for the technical terms used in the claim set. The definitions should not be understood and limiting to the disclosure, but rather as hinting on some ways of understanding implementations of the invention while not excluding interpretations not mentioned in the definitions. Accordingly, other implementations are within the scope of the following paragraphs.

[0012] As used herein, the term "first guidance curve" may refer to an initial speed profile or trajectory calculated by an automated train operation (ATO) system for controlling the movement of a train. This curve may represent a planned sequence of states, like speeds or accelerations over distance or time, taking into account various factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The first guidance curve may serve as a reference for the train's operation until conditions warrant the calculation of a new guidance curve.

[0013] The term "train" as used in the claims refers to any rail vehicle capable of traveling on rails. This includes, but is not limited to, passenger trains, freight trains, locomotives, trams, metros, highspeed trains, and cargo trains. A train may consist of one or more connected rail vehicles, such as a locomotive pulling passenger cars or freight cars. In the context of automated train operation, the term "train" encompasses the entire rail vehicle unit that is being controlled as a single entity.

[0014] The term "rail" refers to the tracks on which the train travels. This includes standard railway tracks, as well as specialized tracks for trams, metros, or other rail-based transportation systems. The rail provides guidance and support for the train's wheels, allowing for controlled movement along a predetermined path.

[0015] "Automated train operation" (ATO) refers to a system or method for controlling the movement of a train with reduced or no human intervention. ATO systems are typically implemented as software modules installed on an on-board device within the train itself, ensuring real-time control and responsiveness to local conditions. However, these systems can also be implemented on a server or in a cloud environment. A server in this context refers to a computer system dedicated to running specific software applications and services, while a cloud environment refers to a network of remote servers hosted on the internet to store, manage, and process data. The on-board implementation is often preferred for its reliability and low-latency performance, particularly in areas with limited network connectivity. It's important to note that ATO components can also be implemented in hardware, such as dedicated computer systems with processors, memory, and data storage devices specifically designed for train control applications, whether located on-board the train or in centralized control centers.

[0016] A "guidance curve" in the context of ATO refers to a calculated trajectory or speed profile that the train should follow. This curve considers various factors such as track geometry, speed limits, energy efficiency, and passenger comfort. The guidance curve is typically represented as a set of data points defining the desired speed of the train at various positions along its route.

[0017] The term "boundary conditions" encompasses various constraints and parameters that affect the calculation of the guidance curve. These may include speed limits, track gradients, curves, station locations, power limits, and scheduled arrival times. Changes in boundary conditions could involve updates to any of these parameters that would necessitate recalculation of the guidance curve.

[0018] "Deviation" refers to the difference between the train's actual state, e.g. position, time or speed, and the state prescribed by the current guidance curve. This deviation is continuously or at least periodically monitored to determine if recalculation of the guidance curve is necessary.

[0019] The "initial interval" is a period of time or distance immediately following the decision to recalculate the guidance curve, during which the train continues to follow a predetermined intermediate guidance curve. This interval allows time for the new guidance curve to be calculated without requiring an immediate change in the train's behavior.

[0020] The "intermediate guidance curve" is a temporary trajectory or speed profile that the train follows during the initial interval. This could be based on the existing first guidance curve, a fixed value (such as a constant speed or acceleration), or another predefined profile that doesn't need much computational effort.

[0021] In the context of the method claims, each step can be considered as a module in a software or hardware implementation. For example, the step of calculating a guidance curve could be performed by a calculation module, which transforms input data (such as boundary conditions and train characteristics) into output data (the guidance curve). This output data would then be sent as a digital data telegram to a control module by means of a communication module. The control module would receive this data through another communication module and use it to control the train's movement.

[0022] The "target position" refers to a specific location along the rail that the train is intended to reach. This could be a station stop, a point where a speed change is required, or any other significant position along the train's route. The ATO system aims to control the train's movement to arrive at this target position according to the calculated guidance curve.Summary of Invention

[0023] The invention as described in claim 1 provides several significant advantages over previous solutions for controlling trains using automated train operation (ATO): 1. Improved computational efficiency: By calculating a new guidance curve that starts after an initial interval at a future point, the method allows for more efficient use of onboard computing resources. This approach reduces the strain on the ATO system's processor, as it does not need to immediately implement a newly calculated curve. This is particularly advantageous for embedded systems with limited computational power. 2. Continuous operation during recalculation: The use of a predetermined intermediate guidance curve during the initial interval ensures that the train continues to operate smoothly while a new guidance curve is being calculated. This eliminates any potential gaps or interruptions in train control that might occur if the system had to wait for a new curve to be fully computed before continuing operation. 3. Adaptive responsiveness: The method allows for dynamic adjustment of the train's trajectory by recalculating the guidance curve when deviations exceed a predetermined threshold or when boundary conditions change significantly. This ensures that the ATO system can respond effectively to changing conditions while avoiding unnecessary recalculations for minor fluctuations. 4. Smooth transitions: By starting the new guidance curve at a future point after an initial interval, the method allows for smoother transitions between guidance curves, i.e. from the first guidance curve to the new guidance curve. This can lead to improved passenger comfort and reduced wear on train components by avoiding abrupt changes in speed or acceleration. The future point is set taking the predetermined intermediate guidance curve into account, ensuring a seamless transition from the intermediate curve to the new guidance curve. 5. Flexibility in handling boundary condition changes: The method explicitly accounts for changes in boundary conditions, allowing the ATO system to adapt to updates in track conditions, speed limits, or other operational parameters. This ensures that the train always operates according to the most current information available. 6. Balance between responsiveness and stability: The use of predetermined thresholds for deviation and boundary condition changes strikes a balance between being responsive to significant changes and maintaining stable operation. This helps prevent oscillations or frequent unnecessary recalculations that could result from an overly sensitive system. 7. Proactive control: By calculating a new guidance curve that starts at a future point, the method allows the ATO system to anticipate and prepare for upcoming changes in the train's trajectory. This proactive approach can lead to more optimal control and potentially improved energy efficiency. 8. Scalability: The method is applicable to various types of rail vehicles and track configurations, as it does not rely on specific hardware or infrastructure beyond basic ATO capabilities. This makes it a versatile solution that can be implemented across different rail systems. 9. Real-time adaptation: The continuous monitoring of the train's actual state, e.g. position or time, and its deviation from the current guidance curve allows for real-time adaptation to unexpected situations, such as slight delays or variations in train performance. 10. Reduced reliance on frequent communication: By enabling the onboard ATO system to recalculate guidance curves independently, the method reduces the need for frequent communication with centralized control systems. This can be particularly advantageous in areas with limited or unreliable communication infrastructure.

[0024] These advantages collectively contribute to a more robust, efficient, and adaptable ATO system that can improve the overall performance and reliability of automated train operations.Description of Embodiments

[0025] In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to the first guidance curve. This approach offers several advantages. By basing the intermediate guidance curve on the existing first guidance curve, the method ensures a smooth transition during the initial interval while the new guidance curve is being calculated. This continuity helps maintain passenger comfort and reduces unnecessary wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve as a basis for the intermediate curve leverages the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period of the initial interval. Alternative implementations could involve interpolating between key points of the first guidance curve to generate a simplified intermediate curve, or using a subset of the first guidance curve's data points to create a less computationally intensive version for the initial interval.

[0026] In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to a fixed value, for example fixed velocity or fixed acceleration. E.g. the fixed velocity could be the actual velocity at the time when the need for the new guidance curve is derived. This approach offers simplicity and predictability during the initial interval. By maintaining a constant speed or acceleration, the train's behavior becomes highly predictable, which can be advantageous for energy management and passenger comfort. This method also reduces computational requirements during the initial interval, as maintaining a fixed value is less demanding than following a complex curve. Alternative implementations could involve using different fixed values based on the current section of track (e.g., higher fixed speeds on straight sections, lower fixed speeds on curves), or gradually transitioning between fixed values to create a stepped intermediate guidance curve.

[0027] In a further preferred embodiment of the inventive computer-implemented method, the length of the initial interval is set according to at least one current state of the train or at least one boundary condition. This adaptive approach to determining the initial interval length offers increased flexibility and responsiveness to real-time conditions. By considering the train's current state (such as speed, acceleration, or position) or relevant boundary conditions (such as upcoming track geometry or speed restrictions), the method can optimize the length of the initial interval to balance computational needs with operational efficiency. For example, a longer initial interval might be set when approaching a complex track section that requires more extensive calculations for the new guidance curve. Alternative implementations could involve using machine learning algorithms to dynamically adjust the initial interval length based on historical performance data, or implementing a multi-factor scoring system that weighs various states and conditions to determine the optimal interval length.

[0028] In a further preferred embodiment of the inventive computer-implemented method, the length of the initial interval is set according to an estimated time for calculating the new guidance curve. This approach directly links the initial interval to the computational requirements of generating the new guidance curve, ensuring that the train continues to operate smoothly while calculations are in progress. By estimating the calculation time, the method can adapt to varying computational loads or complexities in different scenarios, potentially improving overall system responsiveness. This could be particularly advantageous when dealing with sections of track that require more complex calculations due to multiple boundary condition changes or intricate geometries. Alternative implementations might involve using a rolling average of recent calculation times to estimate future needs, or implementing a predictive model that considers factors like upcoming track complexity and current computational load to estimate calculation times more accurately.

[0029] In a further preferred embodiment of the inventive computer-implemented method, the threshold for the change in the boundary conditions is zero in order to calculate the new guidance curve for every change in the boundary conditions. This approach ensures maximum responsiveness to any alterations in the operational environment. By recalculating the guidance curve for every detected change, no matter how small, the method can provide the most up-to-date and optimized trajectory for the train at all times. This could be particularly beneficial in complex rail environments with frequently changing conditions, such as dense urban networks or areas with variable speed restrictions. However, this approach may increase computational load and could potentially lead to more frequent transitions between guidance curves. Alternative implementations could involve setting non-zero thresholds for different types of boundary conditions based on their relative impact on train operation, or implementing a cumulative change threshold that triggers recalculation only when the sum of multiple small changes exceeds a certain value.

[0030] In a further preferred embodiment of the inventive computer-implemented method, the threshold for the deviation is dynamically adjusted based on at least one of: current speed of the train, distance to the target position, or complexity of upcoming track sections. This adaptive threshold approach allows for more nuanced control over when recalculations are triggered. By considering factors like current speed, remaining distance, or track complexity, the method can balance the need for accuracy with computational efficiency. For example, tighter deviation thresholds could be set when approaching the target position or when navigating complex track sections, while allowing for larger deviations at high speeds on straight sections where minor variations have less impact. Alternative implementations could involve using a multi-dimensional threshold that considers multiple factors simultaneously, or implementing a predictive model that adjusts thresholds based on anticipated future conditions along the route.

[0031] In a further preferred embodiment of the inventive computer-implemented method, the predetermined threshold for the deviation is a function of the distance to the target position. This approach recognizes that the importance of precise adherence to the guidance curve typically increases as the train approaches its target position. By making the deviation threshold a function of remaining distance, the method can become increasingly stringent in its control as the train nears its destination. This can help ensure accurate stopping at platforms or precise positioning at critical points along the route. For example, the threshold could decrease linearly or exponentially as the target position approaches. Alternative implementations might involve using a step function with discrete threshold changes at key distances, or combining the distance-based function with other factors like current speed or track geometry to create a more complex, multi-variable threshold function.

[0032] In a further preferred embodiment of the inventive computer-implemented method, the method further comprises adjusting the speed of the train during the initial interval to facilitate a smooth transition to the new guidance curve. This additional step enhances the overall smoothness of the train's operation during guidance curve transitions. By proactively adjusting the train's speed during the initial interval, the method can minimize potential discontinuities when switching to the new guidance curve. This can lead to improved passenger comfort, reduced wear on train components, and potentially better energy efficiency by avoiding sudden changes in power demand. The speed adjustment could be implemented as a gradual acceleration or deceleration to better align with the anticipated starting point of the new guidance curve. Alternative implementations could involve using predictive algorithms to estimate the likely characteristics of the new guidance curve and adjusting speed accordingly, or implementing a multi-stage transition process with increasingly fine-tuned speed adjustments as the end of the initial interval approaches.

[0033] Based on known devices for train control, the invention further addresses the challenge of providing an arrangement that enables efficient and accurate calculation of guidance curves for autonomous rail vehicles while maintaining smooth operation and reducing computational load. The invention solves this problem through an ATO device according to claim 10. Preferred embodiments result from the dependent claims 11 to 14, whereby the same advantages as initially explained for the inventive computer-implemented method apply analogously.

[0034] Based on known computer program products for calculation of guidance curves, the invention further addresses the challenge of providing a computer program product that enables efficient and accurate calculation of guidance curves for autonomous rail vehicles while maintaining smooth operation and reducing computational load.

[0035] The invention solves this problem through a computer program product according to claim 15. The same advantages as initially explained for the inventive computer-implemented method apply analogously.

[0036] Non-limiting and non-exhaustive examples are described with reference to the following Figures.Brief Description of the Drawings

[0037] For better explanation of the invention, the following schematic representations show: Fig. 1an exemplary embodiment of a rail vehicle system according to the invention, Fig. 2a velocity-distance graph showing guidance curves according to an embodiment of the invention, Fig. 3another alternative velocity-distance graph showing alternative guidance curves according to an embodiment of the invention, and Fig. 4a flowchart of an automated train operation process according to an embodiment of the invention performed by the train in Figure 1. Description of Examples

[0038] Figure 1 illustrates an exemplary embodiment of a rail vehicle system according to the invention. The system comprises a train 1 consisting of a locomotive 2 and a passenger car 3 traveling on a rail 4. The train 1 is moving towards a target position 5 as indicated by a direction indicator 6.

[0039] The locomotive 2 is equipped with a train control system 9 comprising an ATP device 7 and an ATO device 8. The ATO device 8 is configured according to an exemplary embodiment of the invention. The train control system 9 is positioned inside of the locomotive 2. The passenger car 3 is coupled to the locomotive 2, representing a typical configuration for passenger rail transport. The rail 4 provides the track on which the train 1 operates, guiding its movement towards the target position 5.

[0040] The target position 5 is marked on the rail 4, indicating a specific point the train 1 is approaching. The target position 5 may represent a station stop, speed restriction point, or other significant location for the train's operation. The direction indicator 6 shows the train's direction of travel, which is towards the target position 5.

[0041] The ATO device 8 is configured to calculate a first guidance curve 10 for automated train operation taking boundary conditions into account. These boundary conditions may include track geometry, speed limits, and other operational parameters. The ATO device 8 controls the automated driving of the train 1 according to this first guidance curve 10. The calculation of guidance curves according to the invention is described later in more detail.

[0042] The ATO device 8 may comprise a computer on which the inventive computer-implemented method is performed, allowing for real-time processing and execution of the automated train operation algorithms. This computer may include processors, memory, and storage components necessary to carry out the complex calculations and decision-making processes involved in generating and adjusting guidance curves.

[0043] The train control system 9 is capable of deriving information about boundary conditions like actual information points on the route based on stored route information and position information about the current position of the train 1. This capability allows the system to anticipate and respond to upcoming track features or operational requirements.

[0044] To obtain accurate position information, the train control system 9 may have an interface to additional systems of the train 1. These additional systems may include for example an odometer system, a Balise Transmission Module (BTM), or a Global Navigation Satellite System (GNSS) device. By integrating data from these sources, the train control system 9 can maintain precise knowledge of the train's actual position 11 along the rail 4.

[0045] The ATP device 7 works in conjunction with the ATO device 8 to ensure safe operation of the train 1. While the ATO device 8 manages the automated driving functions, the ATP device 7 provides an additional layer of safety oversight, monitoring the train's speed and position to prevent unsafe conditions.

[0046] Figure 4 illustrates a flowchart of a method for controlling the train 1 using automated train operation (ATO). The method is performed by the ATO device 8 of the train 1 introduced in Figure 1. Some parts are shown in Figures 2 and 3, which will be described in more detail below.

[0047] The process begins with a method step 100 of calculating a first guidance curve 10 considering boundary conditions. These boundary conditions may include track geometry, speed limits, and other operational parameters relevant to the train 1 traveling on the rail 4 towards the target position 5.

[0048] The next method step 110 involves controlling train driving according to the first guidance curve 10. The ATO device 8 uses this first guidance curve 10 to manage the automated operation of the train 1.

[0049] The process then moves to a method step 120 of determining an actual state like a position 11 of the train 1, a deviation 12 between the actual state and the first guidance curve 10, and any change in the boundary conditions. This step allows the ATO device 8 to monitor the train's progress and identify any significant deviations from the planned route according to the first guidance curve 10 or any changes in operational conditions. As this step is done continuously or at least periodically, the actual position 11 is shown in Figures 2 and 3 as trajectories. The deviation 12 is shown in the examples in Figures 2 and 3 as a deviation between positions, i.e. the actual position 11 and the position in the first guidance curve 10. However, in alternative embodiments the deviation could also be in time or speed.

[0050] A decision point 160 follows, where the process checks if the deviation 12 or boundary condition change exceeds a predetermined threshold respectively. If no threshold is exceeded, the process loops back to method step 110, continuing to control the train 1 according to the first guidance curve 10.

[0051] If any of the thresholds is exceeded, the process proceeds to a method step 130 to calculate a new guidance curve 15. This new guidance curve 15 starts after an initial interval 16 at a future point 14, allowing time for the calculations to be completed while the train 1 continues to operate. The future point 14 can be precisely determined since the intermediate guidance curve 17 is known, allowing the ATO device 8 to calculate the new guidance curve 15 to start exactly at this predetermined future point 14.

[0052] The length of the initial interval 16 may be set dynamically based on various factors to optimize the train's operation and ensure smooth transitions between the first guidance curve 10 and new guidance curve 15. For example, the length of the initial interval 16 may be determined by considering the current operational state of the train 1, including its speed, position, and acceleration. For instance, a longer initial interval 16 may be set when the train 1 is traveling at higher speeds to allow more time for calculations and gradual adjustments.

[0053] In other implementations, the initial interval 16 may be influenced by the complexity of the upcoming track section. If the train 1 is approaching a section with multiple curves, gradient changes, or speed restrictions, the ATO device 8 may set a longer initial interval 16 to ensure sufficient time for calculating a more complex new guidance curve 15. Conversely, for simpler track sections, a shorter initial interval 16 may be adequate.

[0054] The ATO device 8 may also consider the computational resources available when setting the initial interval 16. If the device is handling multiple tasks simultaneously or operating in a resource-constrained environment, it may extend the initial interval 16 to ensure adequate time for calculations without compromising other critical functions.

[0055] In some cases, the initial interval 16 may be adjusted based on historical data and machine learning algorithms. The ATO device 8 may analyze past performance and optimize the length of the initial interval 16 for different scenarios, potentially improving efficiency over time as it learns from previous operations.

[0056] The system may also consider external factors when setting the initial interval 16. For example, if the train 1 is operating in an area with known communication challenges or intermittent data updates, a longer initial interval 16 may be set to accommodate potential delays in receiving updated boundary conditions or other critical information.

[0057] In certain implementations, the initial interval 16 may be set as a function of the magnitude of the deviation 12 or the extent of changes in boundary conditions. Larger deviations or more significant changes may warrant longer initial intervals to allow for more comprehensive recalculations and smoother transitions.

[0058] The ATO device 8 may also implement adaptive strategies for setting the initial interval 16. It may start with a default interval length and dynamically adjust it based on the success of previous transitions between guidance curves. If smooth transitions are consistently achieved with shorter intervals, the system may gradually reduce the interval length to improve responsiveness. Conversely, if transitions are causing discomfort or operational issues, the system may increase the interval length to allow for more gradual adjustments.

[0059] The next method step 140 involves controlling the train 1 during the initial interval 16 using a predetermined intermediate guidance curve 17. This ensures continuous and smooth operation of the train 1 while the new guidance curve 15 is being calculated and the train 1 reaches the future point 14.

[0060] The intermediate guidance curve 17 may be set using various approaches to ensure smooth operation during the initial interval 16. For example, the ATO device 8 may set the intermediate guidance curve 17 based on the first guidance curve 10, potentially using a simplified version or subset of data points from the first guidance curve 10. This approach may provide continuity in the train's operation while reducing computational load during the transition period. In other implementations, the intermediate guidance curve 17 may be set according to a fixed value, such as a constant velocity or acceleration. This method may be particularly useful in situations where maintaining a steady state is preferable, for example, when approaching a station or navigating a complex track section. The ATO device 8 may select an appropriate fixed value based on the train's current speed, upcoming track conditions, or energy efficiency considerations. The ATO device 8 may also employ more sophisticated methods to set the intermediate guidance curve 17. For instance, it may use a predictive algorithm that considers the train's current state, upcoming track geometry, and anticipated characteristics of the new guidance curve 15 to generate an optimized intermediate trajectory. Anyway, the intermediate guidance curve 17 links to the future point 18, so that the start of the new guidance curve 15 is known.

[0061] Following this, the process moves to a method step 150 of controlling the train 1 starting at the future point 14 using the new guidance curve 15. This allows the ATO device 8 to implement the updated new guidance curve 15 based on the most recent calculations and conditions.

[0062] After this step, the process loops back to method step 120 as described above, continuing the cycle of monitoring, evaluating, and adjusting the train's operation as needed. Note that, after looping back to step 120, the new guidance curve 15 will be handled as the first guidance curve 10 before.

[0063] The flowchart in Figure 4 highlights the iterative nature of the method, with continuous monitoring and adjustment of the guidance curve based on real-time conditions. The process incorporates an adaptive mechanism that allows for recalculation of the guidance curve when significant deviations or changes in boundary conditions occur, while maintaining control during the recalculation period using the intermediate guidance curve 17.

[0064] This method, as performed by the ATO device 8, works in conjunction with the ATP device 7 and other components of the train control system 9 to ensure safe and efficient operation of the train 1 as it travels along the rail 4 towards the target position 5.

[0065] A computer program product may comprise instructions that, when executed by a computer, cause the computer to perform this method for controlling the train 1 using automated train operation. This computer program product may be integrated into the ATO device 8 or other suitable components of the train control system 9.

[0066] Figures 2 and 3 illustrate exemplary velocity-distance graphs showing guidance curves calculated and used by the ATO device 8 of the train 1 introduced in Figure 1. These graphs represent alternative outcomes when applying the method shown in Figure 4.

[0067] Figure 2 depicts a scenario where the new guidance curve 15 is calculated when the deviation 12 exceeds the predetermined threshold. The graph shows the first guidance curve 10 representing the planned speed of the train 1 over distance. The actual position 11 curve shows the real speed of the train 1 over distance. At the first position 13, the deviation 12 between the actual position 11 and the first guidance curve 10 exceeds the predetermined threshold. Consequently, as described above the ATO device 8 calculates the new guidance curve 15.

[0068] The threshold for the deviation 12 may be dynamically adjusted based on factors such as the current speed of the train 1, the distance to the target position 5, or the complexity of upcoming track sections. This adaptive approach allows for more precise control in critical areas while allowing for greater flexibility in less critical sections. The predetermined threshold for the deviation 12 may be set as a function of the distance to the target position 5. This approach typically results in tighter control as the train 1 approaches its destination.

[0069] The new guidance curve 15 starts at the future point 14, which is in the future relative to the first position 13. The distance between the first position 13 and the future point 14 represents the initial interval 16. During this initial interval 16, the ATO device 8 calculates the new guidance curve 15 while controlling the train 1 using the intermediate guidance curve 17.

[0070] In the exemplary situation shown in Figure 2 the intermediate guidance curve 17 is set to a constant velocity, which is the velocity at the first position 13. Here, the future point 14 may be calculated with a high degree of accuracy, as the velocity during the initial interval 16 is fixed. This fixed velocity allows for a precise determination of the train's position at the end of the initial interval 16. The ATO device 8 may use this information to set the starting point for the new guidance curve 15, ensuring that it begins exactly at the future point 14. By calculating the new guidance curve 15 to start at this predetermined future point 14, the system can provide a seamless transition from the intermediate guidance curve 17 to the new guidance curve 15. This approach may allow for smoother train operation and more efficient use of computational resources, as the ATO device 8 can focus on calculating the optimal trajectory from the known future starting point, which is the future point 14.

[0071] Alternatively, the intermediate guidance curve 17 may be set to a constant acceleration if this is deemed appropriate for the current operational context. This approach may be particularly useful in situations where maintaining a constant velocity is not ideal, such as when the train 1 is on an incline or decline, or when energy efficiency considerations favor a gradual change in speed. By setting a constant acceleration, the ATO device 8 may be able to more smoothly transition between the first guidance curve 10 and the new guidance curve 15, potentially improving passenger comfort and reducing wear on the train's components.

[0072] The ATO device 8 may calculate the intermediate guidance curve 17 based on various factors and constraints. In some implementations, the device may analyze the current state of the train 1, including its speed, position, and acceleration, as well as upcoming track conditions and operational requirements. The ATO device 8 may then use this information to determine the most appropriate parameters for the intermediate guidance curve 17, whether it be a constant velocity, constant acceleration, or a more complex profile. The calculation may also consider the estimated duration of the initial interval 16, ensuring that the intermediate guidance curve 17 provides a suitable trajectory for the train 1 until the new guidance curve 15 can be implemented at the second position point 14. Another alternative approach in getting the intermediate guidance curve 17 is now described in relation to Figure 3.

[0073] Figure 3 illustrates an alternative scenario where the new guidance curve 15 is calculated due to a change in a boundary condition 18 exceeding a predetermined threshold. The graph shows the first guidance curve 10 starting from a starting position S0 and increasing in velocity. At a first position 13 at position S1, there is a deviation 12 between the actual position 11 and the first guidance curve 10. But the deviation 12 is not above the threshold. Therefore, this is not the reason for calculation a new guidance curve 15. But at the first position 13, one of the boundary conditions 18 changes, specifically the speed limit Vmax that changes from Vmax to Vmax'.

[0074] In response to this change, the ATO device 8 calculates the new guidance curve 15. The new guidance curve 15 begins at a future point 14 at position S2, after the initial interval 16 between S1 and S2. The new guidance curve 15 reaches a plateau at the new speed limit Vmax' and then decreases as the train 1 approaches the target position 5.

[0075] The length of the initial interval 16 may be set according to an estimated time for calculating the new guidance curve 15. This allows the ATO device 8 to complete the necessary calculations before implementing the new guidance curve 15.

[0076] In some implementations, the threshold for changes in the boundary conditions 18 may be set to zero. This approach ensures that the ATO device 8 calculates a new guidance curve 15 for every change in the boundary conditions 18, no matter how small.

[0077] The intermediate guidance curve 17 used during the initial interval 16 in the exemplary embodiment in Figure 3 is set by the ATO device 8 according to the first guidance curve 10. This approach may offer several advantages. By basing the intermediate guidance curve 17 on the existing first guidance curve 10, the method may ensure a smooth transition during the initial interval 16 while the new guidance curve 15 is being calculated. This continuity may help maintain passenger comfort and reduce wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve 10 as a basis for the intermediate curve 17 may leverage the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period. If necessary, the ATO device 8 may interpolate between key points of the first guidance curve 10 to generate a simplified intermediate curve 17, or use a subset of the first guidance curve's data points to create a less computationally intensive version for the initial interval 16.

[0078] The length of the initial interval 16 may be set according to at least one current state of the train 1 or at least one boundary condition 18. This allows the ATO device 8 to adapt the calculation time based on the current operational context.

Examples

Embodiment Construction

[0025]In a further preferred embodiment of the inventive computer-implemented method, the intermediate guidance curve is set according to the first guidance curve. This approach offers several advantages. By basing the intermediate guidance curve on the existing first guidance curve, the method ensures a smooth transition during the initial interval while the new guidance curve is being calculated. This continuity helps maintain passenger comfort and reduces unnecessary wear on train components that could result from abrupt changes in speed or acceleration. Additionally, using the first guidance curve as a basis for the intermediate curve leverages the existing calculations and boundary conditions, potentially reducing computational overhead during the transition period of the initial interval. Alternative implementations could involve interpolating between key points of the first guidance curve to generate a simplified intermediate curve, or using a subset of the first guidance cur...

Claims

1. A computer-implemented method for controlling a train (1) traveling on a rail (4) towards a target position (5) using automated train operation (ATO), the method comprising: calculating a first guidance curve (10) for automated train operation considering boundary conditions; controlling the driving of the train (1) according to the first guidance curve (10); determining an actual state of the train (1), a deviation (12) between the actual state and the first guidance curve (10), and any change in the boundary conditions; calculating a new guidance curve (15) when the deviation (12) exceeds a predetermined threshold or the change in the boundary conditions exceeds a predetermined threshold, wherein the new guidance curve (15) starts after an initial interval (16) at a future point (14); controlling the driving of the train (1) starting at the future point (14) according to the new guidance curve (15); and controlling the driving of the train (1) during the initial interval (16) according to a predetermined intermediate guidance curve (17).

2. The computer-implemented method of claim 1, characterized in that the intermediate guidance curve (17) is set according to the first guidance curve (10).

3. The computer-implemented method of claim 1 or 2, characterized in that the intermediate guidance curve (17) is set according to a fixed value, for example fixed velocity or fixed acceleration.

4. The computer-implemented method of any of claims 1 to 3, characterized in that the length of the initial interval (16) is set according to at least one current state of the train (1) or at least one boundary condition.

5. The computer-implemented method of any of claims 1 to 4, characterized in that the length of the initial interval (16) is set according to an estimated time for calculating the new guidance curve (15).

6. The computer-implemented method of any of claims 1 to 5, characterized in that the threshold for the change in the boundary conditions is zero in order to calculate the new guidance curve (15) for every change in the boundary conditions.

7. The computer-implemented method of any of claims 1 to 6, characterized in that the threshold for the deviation (12) is dynamically adjusted based on at least one of: current speed of the train (1), distance to the target position (5), or complexity of upcoming track sections.

8. The computer-implemented method of any of claims 1 to 7, characterized in that the predetermined threshold for the deviation (12) is a function of the distance to the target position (5).

9. The computer-implemented method of any of claims 1 to 8, characterized in that the method further comprises adjusting the speed of the train (1) during the initial interval (16) to facilitate a smooth transition to the new guidance curve (15).

10. An automated train operation (ATO) device for controlling a train (1) traveling on a rail (4) towards a target position (5), the ATO device configured to: calculate a first guidance curve (10) for automated train operation considering boundary conditions; control the driving of the train (1) according to the first guidance curve (10); determine an actual state of the train (1), a deviation (12) between the actual state and the first guidance curve (10), and any change in the boundary conditions; calculate a new guidance curve (15) when the deviation (12) exceeds a predetermined threshold or the change in the boundary conditions exceeds a predetermined threshold, wherein the new guidance curve (15) starts after an initial interval (16) at a future point (14); control the driving of the train (1) starting at the future point (14) according to the new guidance curve (15); and control the driving of the train (1) during the initial interval (16) according to a predetermined intermediate guidance curve (17).

11. The automated train operation (ATO) device of claim 10, characterized in that the intermediate guidance curve (17) is set according to the first guidance curve (10).

12. The automated train operation (ATO) device of claim 10 or 11, characterized in that the intermediate guidance curve (17) is set according to a fixed value, for example fixed velocity or fixed acceleration.

13. The automated train operation (ATO) device of any of claims 10 to 12, characterized in that the length of the initial interval (16) is set according to at least one current state of the train (1) or at least one boundary condition.

14. The automated train operation (ATO) device of any of claims 10 to 13, characterized in that the length of the initial interval (16) is set according to an estimated time for calculating the new guidance curve (15).

15. Computer program product comprising instructions that, when executed by a computer, cause the computer to perform a method according to any one of claims 1 to 9.

Images