Rail vehicle warning system

Abstract

The present invention discloses a rail vehicle alert system, which belongs to the technical field of rail transit driver driving status, and includes an alert time module, an alert test module and a server; the alert time module is used to set the alert time interval, and when the alert detection is reached, an alert signal is generated and sent to the alert test module, the alert test module performs alert detection on the rail vehicle driver, establishes an action library, identifies the current operation action of the rail vehicle driver, matches the corresponding detection action set from the action library according to the identified operation action, selects one detection action from the detection action set as an assessment action, sets a display unit, inputs the assessment action into the display unit, issues an alert prompt, and prompts the rail vehicle driver to perform alert detection according to the assessment action, sets a detection time, collects the action of the rail vehicle driver within the detection time, obtains a comparison gesture, and establishes a virtual unit.

Description

Technical Field

[0001] The present invention belongs to the technical field of rail transit driver driving state monitoring, in particular to a rail vehicle alerting system. Background Art

[0002] Rail transit refers to a type of vehicle or transportation system in which operating vehicles need to travel on specific tracks. The most typical rail transit is a railway system consisting of traditional trains and standard railways. With the diversified development of train and railway technology, rail transit has shown an increasing number of types. It is not only used in long-distance land transportation, but also widely used in medium and short-distance urban public transportation. With the rapid development of rail transit, the safety requirements for rail transit are also becoming increasingly higher. As a key focus in rail transit safety, rail vehicle drivers need to ensure that their driving status is competent for rail vehicle driving. Therefore, there is an urgent need for a rail vehicle warning system to determine whether the driving status of rail vehicle drivers is competent for rail vehicle driving and ensure the safe operation of rail vehicles. Summary of the Invention

[0003] In order to solve the problems existing in the above-mentioned solutions, the present invention provides a rail vehicle warning system.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] The rail vehicle alert system includes an alert time module, an alert test module, and a server. The alert time module is used to set the alert time interval. When the alert detection period is reached, an alert signal is generated and sent to the alert test module. The alert test module performs alert detection on the rail vehicle driver. The specific method includes:

[0006] Establish an action library, identify the current operating actions of the rail vehicle driver, match the corresponding detection action set from the action library according to the identified operating actions, select one detection action from the detection action set as an assessment action, set a display unit, input the assessment action into the display unit, issue a warning prompt, and prompt the rail vehicle driver to perform an alert detection according to the assessment action, set a detection time, collect the actions of the rail vehicle driver within the detection time, obtain a comparison gesture, establish a virtual unit, input the comparison gesture and the assessment action into the virtual unit for comparison, and judge whether the comparison gesture meets the requirements. When the judgment result is that it meets the requirements, the warning detection is completed; when the judgment result is that it does not meet the requirements, the rail vehicle driver is prompted to re-assess the action.

[0007] Furthermore, it also includes a start module, which is used to determine whether the current rail vehicle needs to perform an alert detection. When it is determined that an alert detection is required, a start signal is generated and the start signal is sent to the alert time module and the alert test module, and the alert time module and the alert test module start running.

[0008] Furthermore, the method for determining whether the current rail vehicle needs to be alerted includes:

[0009] The engine status of the rail vehicle is acquired in real time. The engine status includes an on state and an off state. When the acquired engine status is an off state, no operation is performed. When the acquired engine status is an on state, the direction signal of the rail vehicle is acquired. When the direction signal is not acquired, no operation is performed. When the direction signal of the rail vehicle is acquired, the speed of the rail vehicle is detected. When the speed of the rail vehicle is not greater than the threshold value X1, no operation is performed. When the speed of the rail vehicle is greater than the threshold value X1, it is determined that an alert detection is required.

[0010] Furthermore, the method for setting the alert time interval includes:

[0011] The rail vehicle speed is acquired in real time, and a speed-time graph is established. The acquired rail vehicle speed and the corresponding acquisition time are input into the speed-time graph, marked as a vehicle speed graph, and a distance model is established. The vehicle speed graphs during the current alert time and the previous alert time are identified and input into the distance model to obtain the restricted distance during the current alert time. The completed distance during the current alert time is calculated in real time based on the vehicle speed graph. When the completed distance is equal to the restricted distance, it is determined that the alert detection time has been reached.

[0012] Furthermore, another method for the alertness test module to perform alertness detection on the rail vehicle driver includes:

[0013] An alarm device is provided, which includes an alarm button, an electric horn button and a wind horn pedal. When the alarm signal sent by the alarm time module is received, an alarm prompt is issued. When the rail vehicle driver receives the alarm prompt, he presses the alarm button, presses the electric horn button or steps on the wind horn pedal. When the rail vehicle driver completes any one of them within the specified time, the alarm detection is judged to be completed.

[0014] Furthermore, when the alert detection fails, it is determined that the driver is in a state of being unable to drive the rail vehicle, an emergency braking signal is generated, and the emergency braking signal is sent to the server, and the server controls the rail vehicle to perform emergency braking.

[0015] Furthermore, the method of establishing the action library includes:

[0016] All operating actions of a rail vehicle driver are obtained, several gesture actions are set for each operating action, and several gesture actions corresponding to the same operating action are integrated into a detection action set; the detection action set is three-dimensionally virtualized using a virtual reality algorithm, a database is established, several storage nodes are set in the database, and the storage nodes are marked according to the operating actions, the three-dimensionally virtualized detection action set is stored in the corresponding storage node, and the current database is marked as an action library.

[0017] Furthermore, the method for collecting the motion of the rail vehicle driver during the detection time includes:

[0018] A spatial coordinate system is established in the rail vehicle cab, an operation action position table is established based on the spatial coordinate system and all operation actions of the rail vehicle driver, the recognized operation actions are obtained, the corresponding spatial coordinates are obtained from the operation action position table based on the recognized operation actions, a gesture collection frame is set according to the obtained spatial coordinates, a mixed reality unit is set, and the gestures in the gesture collection frame are collected and converted by the mixed reality unit to form a virtual three-dimensional gesture, which is marked as a comparison gesture.

[0019] Compared with the prior art, the beneficial effects of the present invention are: by alerting the rail vehicle driver, it is ensured that the driving state of the rail vehicle driver is competent for driving the rail vehicle, the safe operation of the rail vehicle is guaranteed, and serious safety accidents are avoided; by dynamically planning the restricted distance within the current alert time based on the vehicle speed map within the previous alert time and the vehicle speed map within the current alert time, and comprehensively considering the relationship between time and speed, the planned alert time is made more comprehensive and reasonable; by making full use of virtual reality technology and mixed reality technology, a three-dimensional comparison of comparison gestures and assessment actions is achieved, making the comparison results more intuitive and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is a principle block diagram of the present invention;

[0022] Figure 2 This is an example diagram of vehicle speed according to the present invention. DETAILED DESCRIPTION

[0023] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0024] like Figures 1 to 2 As shown, the rail vehicle alert system includes a start module, an alert time module, an alert test module and a server;

[0025] The startup module is used to determine whether the current rail vehicle needs to be alerted. Because the rail vehicle alert system is not turned on in real time, it needs to meet certain conditions before it is turned on to test the rail vehicle driver. The method for determining whether the current rail vehicle needs to be alerted includes:

[0026] The engine status of the rail vehicle is acquired in real time. The engine status includes an on state and an off state. When the acquired engine status is an off state, no operation is performed. When the acquired engine status is an on state, the direction signal of the rail vehicle is acquired. When the direction signal is not acquired, no operation is performed. When the direction signal of the rail vehicle is acquired, the speed of the rail vehicle is detected. When the speed of the rail vehicle is not greater than the threshold value X1, no operation is performed. Generally, X1 is 1km / h and can be adjusted according to actual conditions. When the speed of the rail vehicle is greater than the threshold value X1, a start signal is generated and the start signal is sent to the alert time module and the alert test module. The alert time module and the alert test module start running.

[0027] The alert time module is used to set the alert time interval. In existing alert time setting methods, some set a standard time interval as the alert time, and trigger the alert operation at the time. However, this method does not take into account the speed of the rail vehicle within the time interval, resulting in an incomplete alert time setting and loopholes that are easily exploited. Another method is to set it based on the maximum speed of the current rail vehicle. For example, if the maximum speed of the current rail vehicle is 100 km / h, assuming that the alert time corresponding to the current speed is 7 minutes, the remaining time of the current alert detection point is determined based on the last alert detection point. Although this method takes the speed issue into account, the setting of the alert time is still not comprehensive. Therefore, it is necessary to optimize the current alert time setting. The specific methods include:

[0028] Establish a speed time diagram, obtain the rail vehicle speed in real time, input the obtained rail vehicle speed and the corresponding acquisition time into the speed time diagram, and mark it as the vehicle speed diagram, such as Figure 2As shown, A1, A2, A3, A4, and A5 are all alert detection points, that is, detection is performed at the alert detection points, and the time period between two alert detection points is the alert time. A distance model is established. The distance model is established based on a CNN network or a DNN network, using the vehicle speed map during the current alert time and the vehicle speed map during the previous alert time as input items and the corresponding set restricted distance as a training set for training. The vehicle speed map during the current alert time is the vehicle speed map from the last alert detection point to the current time. The vehicle speed map during the current alert time and the previous alert time is identified and input into the distance model to obtain the restricted distance during the current alert time. The completed distance during the current alert time is calculated in real time based on the vehicle speed map. When the completed distance equals the restricted distance, an alert signal is generated and sent to the alert test module. The present application dynamically plans the restricted distance during the current alert time based on the vehicle speed map during the previous alert time and the vehicle speed map during the current alert time, comprehensively considering the relationship between time and speed, making the planned alert time more comprehensive and reasonable.

[0029] The alertness test module is used to detect the alertness of rail vehicle drivers;

[0030] In one embodiment, an alarm device is provided, which includes an alarm button, an electric horn button and a wind horn pedal. When the alarm signal sent by the alarm time module is received, an alarm prompt is issued, which is generally a voice prompt. When the rail vehicle driver receives the alarm prompt, he presses the alarm button, presses the electric horn button or steps on the wind horn pedal. The above actions have equivalent effects. As long as the rail vehicle driver completes any one of them within the specified time, it is deemed to complete the alarm detection. The specified time is discussed and set by the expert group, and is generally about 90 seconds.

[0031] In one embodiment, an action library is established. When an alert signal sent by an alert time module is received, the current operating action of the rail vehicle driver is identified. Since how to identify the current operating action of the rail vehicle driver is common knowledge in the field, it will not be described in detail. According to the identified operating action, a corresponding detection action set is matched from the action library. One detection action is selected from the detection action set as an assessment action. A display unit is provided, which is used to display the received assessment action. The assessment action is input into the display unit, and an alert prompt is issued to prompt the rail vehicle driver to perform an alert detection according to the assessment action. A detection time is set, that is, the alert detection of the rail vehicle driver is performed within the detection time. If the time is exceeded, the detection is deemed to have failed. The rail vehicle driver's action is collected within the detection time to obtain a comparison gesture. A virtual unit is established, which is used to display a virtual assessment action and a comparison gesture, and the assessment action and the comparison gesture are compared. The comparison gesture and the assessment action are input into the virtual unit for comparison to determine whether the comparison gesture meets the requirements. When the judgment result is that it meets the requirements, the alert detection is completed; when the judgment result is that it does not meet the requirements, the rail vehicle driver is prompted to re-take the action assessment. By making full use of virtual reality technology and mixed reality technology, a three-dimensional comparison of comparison gestures and assessment actions can be achieved, making the comparison results more intuitive and accurate.

[0032] When the alert detection fails, it is determined that the driver is in a state of being unable to drive the rail vehicle, an emergency braking signal is generated, and the emergency braking signal is sent to the server, and the server controls the rail vehicle to perform emergency braking.

[0033] Methods for building an action library include:

[0034] Acquire all the operating actions of the rail vehicle driver, that is, the operating actions that the rail vehicle driver performs in the process of driving the rail vehicle. Because the rail vehicle driver does not perform many operating actions in the driving process, the expert group manually sets the corresponding gesture action set of the operating actions, that is, sets several gesture actions for each operating action, and integrates several gesture actions corresponding to the same operating action into a detection action set; by performing gesture verification, the driving impact of the rail vehicle driver during the detection process is reduced. For example, when the rail vehicle driver is operating a certain rod, the designed gesture actions are gestures such as "OK" and "scissors". The rail vehicle driver's hand can complete these gesture actions on the operating rod without affecting the rail vehicle driver's driving; use the virtual reality algorithm to three-dimensionally virtualize the detection action set. The virtual reality algorithm is an existing algorithm that can use existing equipment or software to perform three-dimensional virtualization of the detection action set, establish a database, set several storage nodes in the database, and mark the storage nodes according to the operating actions, store the three-dimensionally virtualized detection action set in the corresponding storage nodes, and mark the current database as an action library.

[0035] Methods for collecting motion data of rail vehicle drivers during the detection period include:

[0036] A spatial coordinate system is established in the rail vehicle cab, and an operation action position table is established based on the spatial coordinate system and all the operation actions of the rail vehicle driver. The operation action position table represents the current operation action, and the coordinates at which the rail vehicle driver's hand that needs to make a gesture will be located. Because the operation action can be used to understand what device is being operated, the coordinates at which the vehicle driver's hand that needs to make a gesture will be located can be determined based on the spatial coordinate system, and the recognized operation action is obtained. The corresponding spatial coordinates are obtained in the operation action position table based on the recognized operation action, and a gesture collection frame is set based on the obtained spatial coordinates, which is equivalent to establishing a collection frame with the spatial coordinates as the center, and setting a mixed reality unit. The mixed reality unit is used to superimpose real things on the virtual world. The mixed reality unit is established by utilizing the existing mixed reality algorithm; the gestures in the gesture collection frame are collected and converted by the mixed reality unit to form a virtual three-dimensional gesture, which is marked as a comparison gesture.

[0037] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0038] The working principle of the present invention is as follows: a start module is used to determine whether the current rail vehicle needs to perform an alert test. When it is determined that the alert test is required, a start signal is generated and sent to the alert time module and the alert test module, and the alert time module and the alert test module start running; the alert time module sets the alert time interval. When the alert test is reached, an alert signal is generated and sent to the alert test module. The alert test module performs an alert test on the rail vehicle driver, establishes an action library, identifies the current operation action of the rail vehicle driver, matches a corresponding detection action set from the action library according to the identified operation action, selects one detection action from the detection action set as an assessment action, sets a display unit, inputs the assessment action into the display unit, issues an alert prompt, and prompts the rail vehicle driver to perform an alert test according to the assessment action. A detection time is set, and the rail vehicle driver's action is collected within the detection time to obtain a comparison gesture. A virtual unit is established, and the comparison gesture and the assessment action are input into the virtual unit for comparison to determine whether the comparison gesture meets the requirements. When the judgment result is that the requirements are met, the alert test is completed; when the judgment result is that the requirements are not met, the rail vehicle driver is prompted to re-take the action assessment.

[0039] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The rail vehicle warning system is characterized by: It includes an alert time module, an alert test module and a server; the alert time module is used to set the alert time interval. When the alert detection is reached, an alert signal is generated and sent to the alert test module. The alert test module performs alert detection on the rail vehicle driver. The specific method includes: Establish an action library, identify the current operation action of the rail vehicle driver, match the corresponding detection action set from the action library according to the identified operation action, select one detection action from the detection action set as the assessment action, set a display unit, input the assessment action into the display unit, issue an alert prompt, and prompt the rail vehicle driver to perform an alert detection according to the assessment action, set a detection time, collect the action of the rail vehicle driver within the detection time, obtain a comparison gesture, establish a virtual unit, input the comparison gesture and the assessment action into the virtual unit for comparison, and judge whether the comparison gesture meets the requirements. When the judgment result is that it meets the requirements, the alert detection is completed; when the judgment result is that it does not meet the requirements, the rail vehicle driver is prompted to re-take the action assessment; Methods for setting the alert interval include: The rail vehicle speed is acquired in real time, and a speed-time graph is established. The acquired rail vehicle speed and the corresponding acquisition time are input into the speed-time graph, marked as a vehicle speed graph, and a distance model is established. The vehicle speed graphs during the current alert time and the previous alert time are identified and input into the distance model to obtain the restricted distance during the current alert time. The completed distance during the current alert time is calculated in real time based on the vehicle speed graph. When the completed distance is equal to the restricted distance, it is determined that the alert detection time has been reached.

2. The rail vehicle alert system according to claim 1, characterized in that: It also includes a start module, which is used to determine whether the current rail vehicle needs to perform an alert detection. When it is determined that an alert detection is needed, a start signal is generated and the start signal is sent to the alert time module and the alert test module, and the alert time module and the alert test module start running.

3. The rail vehicle alert system according to claim 2, characterized in that: Methods for determining whether the current rail vehicle needs to be alerted include: The engine status of the rail vehicle is acquired in real time. The engine status includes an on state and an off state. When the acquired engine status is an off state, no operation is performed. When the acquired engine status is an on state, the direction signal of the rail vehicle is acquired. When the direction signal is not acquired, no operation is performed. When the direction signal of the rail vehicle is acquired, the speed of the rail vehicle is detected. When the speed of the rail vehicle is not greater than the threshold value X1, no operation is performed. When the speed of the rail vehicle is greater than the threshold value X1, it is determined that an alert detection is required.

4. The rail vehicle alert system according to claim 1, characterized in that: Another method for the alertness test module to detect the alertness of the rail vehicle driver includes: An alarm device is provided, which includes an alarm button, an electric horn button and a wind horn pedal. When the alarm signal sent by the alarm time module is received, an alarm prompt is issued. When the rail vehicle driver receives the alarm prompt, he presses the alarm button, presses the electric horn button or steps on the wind horn pedal. When the rail vehicle driver completes any one of them within the specified time, the alarm detection is judged to be completed.

5. The rail vehicle alerting system according to claim 1 or 4, characterized in that: When the alert detection fails, it is determined that the driver is in a state of being unable to drive the rail vehicle, an emergency braking signal is generated, and the emergency braking signal is sent to the server, and the server controls the rail vehicle to perform emergency braking.

6. The rail vehicle alerting system according to claim 1, characterized in that: Methods for building an action library include: All operating actions of a rail vehicle driver are obtained, several gesture actions are set for each operating action, and several gesture actions corresponding to the same operating action are integrated into a detection action set; the detection action set is three-dimensionally virtualized using a virtual reality algorithm, a database is established, several storage nodes are set in the database, and the storage nodes are marked according to the operating actions, the three-dimensionally virtualized detection action set is stored in the corresponding storage node, and the current database is marked as an action library.

7. The rail vehicle alerting system according to claim 1, characterized in that: Methods for collecting motion data of rail vehicle drivers during the detection period include: A spatial coordinate system is established in the rail vehicle cab, an operation action position table is established based on the spatial coordinate system and all operation actions of the rail vehicle driver, the recognized operation actions are obtained, the corresponding spatial coordinates are obtained from the operation action position table based on the recognized operation actions, a gesture collection frame is set according to the obtained spatial coordinates, a mixed reality unit is set, and the gestures in the gesture collection frame are collected and converted by the mixed reality unit to form a virtual three-dimensional gesture, which is marked as a comparison gesture.

Images