Rail vehicle with autonomous driving system

ES3081279T3Undetermined Publication Date: 2026-10-09ALSTOM HOLDINGS SA
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
ES2020737049T
Authority / Receiving Office
ES · ES
Patent Type
Patents
Current Assignee / Owner
Filing Date
2020-07-13
Publication Date
2026-10-09
Estimated Expiration
2040-07-13

AI Technical Summary

Technical Problem

Existing technologies do not adequately address the need for autonomous driving systems in trams, particularly on sidings within the network, managing traction and braking based on environmental obstacles.

Method used

A tram equipped with an autonomous driving system that utilizes multiple sensors, including LiDARs and inertial measurement units, to detect obstacles ahead and adjust speed and braking accordingly, using a computing unit to determine appropriate responses.

Benefits of technology

Enables safe and autonomous operation of trams by dynamically adjusting speed and braking to avoid obstacles, ensuring smooth navigation on predefined tracks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000008_0000
    Figure 00000008_0000
  • Figure 00000009_0000
    Figure 00000009_0000
  • Figure 00000010_0000
    Figure 00000010_0000
Patent Text Reader

Abstract

This railway vehicle (1) has a manipulator (6), located in the cab, connected to an input of a control / monitoring unit (8) to operate the vehicle's traction / braking means (3, 4). The manipulator is capable of generating a first setpoint signal applied to an input of the control / monitoring unit, which corresponds to a traction or braking force. It is characterized in that, for autonomous operation of the vehicle, it includes an autonomous driving system (10), capable of generating a second setpoint signal, of the same nature as the first, which is applied to said input of the control / monitoring unit.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to the field of autonomous driving of a railway vehicle, and more particularly of a tramway.

[0002] In general, for driving vehicles (cars, buses, trains, etc.) a reference autonomy scale has been defined which allows us to qualify the degree of autonomy with which the vehicle in question moves.

[0003] In this document, the desired level of autonomy is level "4" on this autonomy scale. At level "4", the vehicle's acceleration, braking, and steering are managed automatically; the monitoring of the environment in which the vehicle is traveling is also managed automatically; and the vehicle's reaction to any problems (such as the presence of an obstacle) is also managed by the vehicle itself. Level "4" differs from the highest level, level "5", in that autonomous driving is limited to certain predefined road sections and not to all roads on which the vehicle might travel.It should be noted that level "0" on this autonomy scale corresponds to the case where all the previous operations are performed by the driver, and level "1" to the case where the on-board system offers simple assistance to the vehicle's driver for acceleration, braking, and steering. Level "1" thus corresponds to driving assistance, commonly referred to as DAS for "Driving Assistance System". Driver Assistance System ».

[0004] Document CN 109 774 749 proposes adding a computer to the on-board component of a Communication Based Train Control (CBTC) system to allow a train to run autonomously when it is not carrying passengers.

[0005] Document DE 10 2016 204 984 discloses a railway vehicle converted to be autonomous. An ATO (Automatic Train Control) is added and connected, via a bus, to input interfaces of various train control units.

[0006] Document AU 2014 406 175 discloses a guided vehicle equipped with an obstacle detection system.

[0007] It is desirable to be able to equip a tram with an autonomous driving system, at least on sidings within the network on which it operates. This involves managing the tram's traction and braking according to its environment.

[0008] The invention therefore aims to solve this problem.

[0009] For this purpose the invention relates to a railway vehicle conforming to the attached claims.

[0010] According to the invention, if the tram's position along the tracks allows for the definition of a reference speed, this speed is modulated to account for the presence of an obstacle only if it is located in front of the tram. The system thus delivers a setpoint speed to the speed control module that manages the tram's traction / braking systems.

[0011] Such a system is therefore based on the ability to detect obstacles ahead of the tram along the track. Once an obstacle is located, the system determines the appropriate response, specifically whether and to what extent the tram's braking system should be activated.

[0012] The invention and its advantages will be better understood upon reading the following detailed description of a particular embodiment, given solely by way of non-limiting example, this description being made with reference to the accompanying drawings in which: there figure 1 is a schematic representation of a tramway equipped with an autonomous driving system according to the invention; the figure 2 is a functional block representation of the autonomous driving system of the figure 1 ; THE figures 3 à 5 schematically represent different areas to monitor, defined in front of the tram, depending on the configuration of the track on which the tram travels. figure 1 ; and, the figure 6 is a schematic representation of an embodiment of the operation of the vehicle according to the invention.

[0013] On the figure 1 , is represented a railway vehicle, in particular a tram 1 running on a track 2.

[0014] In a conventional manner, in order to be operated by a driver, tram 1 includes a control system for the tram's traction / braking mechanisms, which are schematically represented on the figure 1 by a motor 3 and brakes 4.

[0015] The control system includes, in the cab, a joystick 6 which allows the generation of command signals. The joystick 6 has a pivoting handle which, when moved forward, allows the driver to request a traction force and, when pivoted backward, allows the driver to request a braking force.

[0016] For example, and preferably, the manipulator can be moved forward to a first position corresponding to a +50% increase in the pulling force, or a second position corresponding to a +15% increase in the pulling force. It can also be moved backward to a first position corresponding to a +50% increase in the braking force, or a second position corresponding to a +15% increase in the braking force.

[0017] The setpoint signal generated by the manipulator 6 is applied as input to a control unit 8 for the traction / braking means. Unit 8 is specifically designed to control the traction / braking means by applying appropriate control signals generated based on the received setpoint signal.

[0018] To be able to be driven autonomously according to level "4" of the scale commonly used for autonomous vehicle driving, tram 1 is equipped with an autonomous driving system 10.

[0019] System 10 is designed to generate setpoint signals of the same nature as those generated by manipulator 6. In this way, the signals generated by system 10 can be applied directly to the input of the control / command unit 8 in place of the setpoint signals generated by manipulator 6.

[0020] Thus, system 10 uses the existing control channel of unit 8 and unit 8 is unchanged with respect to the prior art.

[0021] To be able to switch from a driver-operated mode to an autonomous driving mode, tram 1 is advantageously equipped with a two-state switch 12, allowing the application to the input of unit 8 of either a setpoint signal from manipulator 6, or a setpoint signal from system 10.

[0022] Advantageously, the state of switch 12 is controlled by a switch 14 located in the cab. Switch 14 is activated by the driver, for example, when tram 1 leaves a service track of the network to enter a siding of the network, a siding on which independent operation is permitted. Activation of switch 14 causes switch 12 to switch.

[0023] System 10 includes a hardware layer represented on the figure 1 .

[0024] The system 10 comprises, as acquisition means, a plurality of sensors connected to the input of a computing unit 20.

[0025] Among the sensors used by the autonomous driving system 10, the tram is equipped with an inertial measurement unit 21, capable of determining an instantaneous vector acceleration of the tram and, by time integration, a first instantaneous speed of the tram.

[0026] The system also uses tachometric means 22, allowing a second instantaneous speed of tram 1 to be determined and, by time integration from a reference point along the track, the distance traveled by the tram from this reference point and consequently a second current position of the tram.

[0027] The sensors also include a sensor for a localization system to determine an initial current position of the tram. Preferably, in the embodiment described here in detail, the localization system sensor is a LiDAR 23. The LiDAR 23 is a two-dimensional LiDAR with a wide field of view, for example, 270°, which, by processing the received echoes, allows the tram's environment to be reconstructed and a reconstructed map to be obtained. The plurality of sensors also includes, particularly for the obstacle detection function, a first LiDAR 25 and a second LiDAR 26.

[0028] The first LIDAR 25 is a three-dimensional LIDAR allowing the observation of a first region located in front of tram 1.

[0029] The second LIDAR 26 is a two-dimensional LIDAR allowing observation of a second region located in front of the tramway.

[0030] The cross-referencing of information delivered by the first and second LIDAR 25 and 26 allows the obstacle detection function to determine the relative position of an obstacle with respect to the front of the tram within an extended observation area.

[0031] With particular reference to the figure 2 , which incorporates the various components of the figure 1 but in a functional form, system 10 will be presented in more detail, in particular the calculation unit 20.

[0032] The computing unit 20 includes a dedicated LIDAR localization module 33 that takes as input the signals delivered by the LIDAR 23 and generates a reconstructed map as output. This reconstructed map is applied as input to a localization module 32. The temporal evolution of the reconstructed environment also provides information on the tram's speed, which is then passed to a tram speed estimation module 34.

[0033] The tram speed estimation module 34, in addition to the speed output from module 33, takes as input the signals delivered, on the one hand, by the inertial unit 21 and, on the other hand, by the tachometric means 22.

[0034] Module 34's function is to provide a precise estimate of the tram's instantaneous speed (V on the figure 1 ).

[0035] This value of the speed delivered at the output of module 34 is applied, on the one hand, to a speed regulation module 38 and, on the other hand, to the localization module 32.

[0036] The localization module 32, in addition to estimating instantaneous speed and reconstructing the map, takes as input information delivered by the inertial unit 21. The localization module 32 has the function of delivering a precise estimate of the instantaneous position of the tram on the tracks.

[0037] In the embodiment presented here in detail, module 32 relies on a map 31 of the region where tram 1 is authorized to operate autonomously. This is, for example, a map of the terrain topology (ground surface, buildings, permanently installed equipment, etc.) over which the depot's sidings extend, within which the tram operates autonomously.

[0038] The localization module 32 compares the environment reconstructed from the states received by the LIDAR 23 with the mapping 31 in order to determine a first current position of the tram.

[0039] This first position is corrected from the information provided by the sensors 21 and the time integration of the speed provided by the module 34.

[0040] The current position of tram 1 is applied to a mission management module 35 and a speed calculation module 37.

[0041] Mission management module 35 is executed when switch 12 is toggled to enable autonomous driving mode.

[0042] Module 35 then acquires the tram's current position to determine its location on the rail network. Based on this information and possibly other information provided by a control center, a mission is assigned to tram 1. This mission might, for example, consist of instructing system 10 whether tram 1 should park.

[0043] Speed ​​calculation module 37 is designed to define a reference speed, based in particular on the track topology on which tram 1 is traveling at the current time. Module 37 therefore defines the reference speed according to the tram's current position and a track plan of the network.

[0044] For example, this reference speed is 10 km / h for a straight section of track and 7 km / h for a curved section of track or a section of track approaching a switch.

[0045] Based on the information provided by the obstacle detection module 39, the speed calculation module weights the reference speed in order to deliver a set speed for tram 1. Indeed, if the obstacle detection module detects an obstacle in front of the tram, the reference speed must be reduced to avoid any collision.

[0046] The set speed, provided by module 37, is applied to the input of speed control module 38. This module, by comparing the set speed with the estimated current tram speed provided by module 34, determines the appropriate traction or braking force. The corresponding set signal, of the same type as that generated by control unit 6, is applied, via switch 12, to the tram's traction / braking control unit 8.

[0047] Module 39 is specifically designed to acquire the signals delivered by the LIDARS 25 and 26.

[0048] Module 39 provides, in the event of the presence of an obstacle within the region observed by the LIDARS 25 and 26, the position of this obstacle relative to the front of the tram.

[0049] The operation of system 10 in the event of obstacle detection by module 39, and more specifically, how the speed calculation module 37 determines a setpoint speed, will now be presented in relation to the figures 3 à 5 and the process of figure 6 .

[0050] Based on the current position of tram 1, module 37 determines the type of track on which the tram is traveling at that moment. To do this, module 37 uses, for example, a track plan. This plan associates each position of the tram with a track type.

[0051] Three types of track are predefined, for example. These three types of track are respectively represented in figures 3 , 4 And 5 .

[0052] The first type corresponds to a straight section of track.

[0053] The second type corresponds to a curved section of track.

[0054] Finally, the third type corresponds to a section of track approaching a switch.

[0055] Other types of track can be defined.

[0056] Recognizing the type of track on which the tram is running allows module 37 to define a reference speed value. For example, the reference speed associated with the first type of track is 10 km / h; with the second type, 7 km / h; and with the third type, 7 km / h.

[0057] Other reference speed values ​​are conceivable.

[0058] Furthermore, module 37 defines, in front of the tramway, a surveillance zone within the observation area of ​​module 39 for obstacle detection.

[0059] The surveillance zone follows the path of the track section in front of the tram. Determining the track type thus allows us to determine the geometry of the surveillance zone.

[0060] This surveillance zone extends laterally on either side of the track so as to correspond approximately to the transverse dimensions of tram line 1.

[0061] The length of the surveillance zone depends on the current speed of the tram.

[0062] The surveillance zone allows filtering of all observations delivered by module 39. Only obstacles whose location falls within the surveillance zone will be retained.

[0063] Therefore, obstacles detected by module 39 but located outside the surveillance zone will have no impact on the tram's operation. This prevents sudden braking as soon as an object is detected in the tram's vicinity, even if it poses no risk of collision.

[0064] As illustrated on the figure 3 The Z1 monitoring zone for the first track type is rectangular, while for the second track type, the Z2 monitoring zone is a portion of a torus. For the third track section type, and to maximize safety, the system does not consider the interlocking state of the switch. Consequently, the Z3 monitoring zone covers both outgoing tracks of the switch.

[0065] A particularly advantageous feature is that the surveillance area is subdivided into domains. For example, three domains are defined within the surveillance area at any given time.

[0066] A first zone, called the deceleration zone, is defined away from the tramway. Then, closer to the tramway and continuous with the deceleration zone, is a second zone, called the braking zone. Finally, between the braking zone and the front end of the tramway, is a third zone, called the emergency braking zone.

[0067] Thus, for example, on the figure 3 If an obstacle O is present within the deceleration zone D13 of zone Z1, module 37 calculates a target speed that will lead to a decrease in the tram's instantaneous speed. This can be achieved passively by using wheel / rail friction forces to gradually reduce the tram's speed, or by applying light braking.

[0068] If an obstacle is present within the braking zone D12, module 37 generates a target speed that brings the tram to a stop at a minimum safety distance in front of the obstacle. This minimum safety distance is, for example, 1.2 m.

[0069] Finally, if an obstacle is detected in the emergency braking range D11, module 37 delivers a set speed leading to emergency braking to stop the tram in the shortest possible distance.

[0070] Similarly, on the figure 4 , the Z2 monitoring zone is subdivided into a slowing area D23, a braking area D23 and an emergency braking area D13.

[0071] On the figure 5 , in the case of a track section approaching a switch, the Z3 monitoring area is subdivided into a pair of deceleration domains D33 and D33' (each of these two domains being associated with one of the two exit tracks of the switch), a pair of braking domains D32 and D32' (each of these two domains being associated with one of the two exit tracks of the switch), and an emergency braking domain D31 (common to both exit tracks of the switch).

[0072] It should be noted that the figure 5 represents the surveillance zone as the tram enters the switch, with the front of the tram being roughly directly above the switch. In fact, the shape of the surveillance zone continuously evolves between that shown on the... figures 3 And 4 and the one depicted on the figure 5 depending on the distance of the tram to the switch.

[0073] To the figure 6 The process implemented by module 37 for calculating the setpoint speed is illustrated in block form.

[0074] At step 110, based on the current position of the tram and a plan of the railway network, module 37 determines the type of track on which tram 1 runs.

[0075] In step 120, this information allows the reference speed to be defined.

[0076] At stage 130, a surveillance zone is defined at the front of the train. Advantageously, this surveillance zone is defined not only according to the type of track on which the tram travels, but also according to the tram's speed. For example, the higher the tram's speed, the longer the surveillance zone.

[0077] The surveillance area is subdivided into three areas, for example according to predefined proportions.

[0078] Finally, in step 140, module 37 determines whether an obstacle is detected within the monitoring area. It uses the obstacle location information determined by module 39 for this purpose.

[0079] If there are no obstacles in the monitoring area, the process loops back on itself and returns to step 110. Process 100 is thus iterated for each sampling period.

[0080] On the other hand, if an obstacle is detected at step 140, at step 150, module 37 identifies the domain in which the obstacle is located.

[0081] At step 160, depending on the domain identified in step 150, module 37 chooses an action to perform from among slowing down the tram, braking the tram or emergency braking.

[0082] At step 170, the corresponding speed setpoint is generated and applied to the speed control module 38.

[0083] If the obstacle was within the deceleration zone, at step 180 the process loops back to step 140 to determine the evolution of the obstacle's position. Either the obstacle has left the monitoring zone and the process restarts at step 110, or the obstacle is still present and steps 150 and subsequent steps are performed again.

[0084] If, at step 140, a braking or emergency braking procedure was initiated, at the end of step 170, the tram is allowed to come to a complete stop before proceeding to step 150, which verifies whether the obstacle has been removed from the track. If so, the tram can restart. The process then loops back to step 110.

Claims

1. Rail vehicle (1) comprising a manipulator (6), in the cabin, connected to an input of a monitoring / control unit (8) for actuating the traction / braking means (3, 4) of the rail vehicle, the manipulator being capable of generating a first setpoint signal applied on an input of the monitoring / control unit (8), the setpoint signal corresponding to a traction or braking force, the rail vehicle (1) comprising, for the autonomous driving of the rail vehicle (1), an autonomous driving system (10), capable of generating a second setpoint signal, of the same nature as the first setpoint signal, to be applied on said input of the monitoring / control unit (8), the autonomous driving system (10) comprises obstacle detection means (39), and the autonomous driving system (10) is capable of defining, in front of the rail vehicle, a monitoring zone within an observation region of the obstacle detection means (39), the autonomous driving system (10) is capable of determining a track type on which the rail vehicle is travelling at the current time, a geometry of the monitoring zone being dependent on the type of track determined, characterised in that the monitoring zone is: for a track type corresponding to a rectilinear or curved track section, a surface covering said track; and, for a track type corresponding to a switch, a surface covering the two tracks exiting the switch.

2. Rail vehicle according to claim 1, comprising a switch (12) which can be controlled so as to toggle between a first position of application of the first setpoint signal on the input of the monitoring / control unit (8) and a second position of application of the second setpoint signal on the input of the monitoring / control unit (8).

3. Rail vehicle according to claim 2, comprising a switch (14) in the cabin allowing the driver to switch the switch from the first position to the second position and vice versa.

4. Rail vehicle according to any one of claims 1 to 3, wherein the autonomous driving system (10) comprises: a first module (34) for computing an estimated speed of the rail vehicle; a second module (37) for computing a setpoint speed from the information delivered by the obstacle detection means (39); and a third speed regulation module (38) capable of comparing the setpoint speed with the estimated speed to determine an adapted traction or braking force and generate the second setpoint signal accordingly.

5. Rail vehicle according to claim 1, wherein the autonomous driving system (10) is capable of subdividing the monitoring zone into three successive sectors according to a relative distance between the obstacle and the front of the vehicle determined by the obstacle detection means, the autonomous driving system (10) generating a second setpoint signal corresponding to a deceleration of the vehicle in a first sector, a braking to stop the vehicle at a minimum distance from the obstacle in a second sector, and an emergency braking to stop the vehicle over the shortest possible distance in a third sector.

6. Rail vehicle according to any one of claims 1 to 4, wherein the manipulator (6) accesses the monitoring / control unit (8) via a channel and in that the autonomous driving system (10) uses said channel to access the monitoring / control unit (8), said monitoring / control unit (8) being unchanged.

7. Rail vehicle according to any one of claims 1 to 6, said vehicle being a tram.