Method for calibrating a traffic monitoring system and traffic monitoring system

The method allows for the precise and automated calibration of sensors in traffic monitoring systems using static objects, addressing the challenge of complex recalibration processes and ensuring continuous system operation and safety.

DE102023213136A1Pending Publication Date: 2025-06-26CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
DE102023213136
Authority / Receiving Office
DE · DE
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-12-21
Publication Date
2025-06-26

AI Technical Summary

Technical Problem

Existing traffic monitoring systems, such as AVP type 2 systems, require complex and time-consuming recalibration processes when sensors fail or become misaligned, necessitating interruption of productive operation.

Method used

A method for calibrating sensors in a traffic monitoring system using static objects within the monitoring area, where the position of these objects is stored in a control unit, allowing for automatic detection and recalibration of sensors without interrupting system operation.

Benefits of technology

Enables precise, automated, and continuous calibration of sensors, ensuring accurate vehicle guidance and collision avoidance without requiring system downtime, thus enhancing operational efficiency and safety.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

The invention relates to a method for calibrating a traffic monitoring system (1). The traffic monitoring system (1) has at least two sensors (3) for detecting a monitoring area (11) and a control unit connected to the sensors (3). At least one static object (5) is provided in the monitoring area (11). The position of the static object (5) is stored in the control unit. The method comprises the following steps: a) detecting the static object (5) with at least one sensor (3) and determining the position of the static object (5) from the sensor information, and b) Calibrating the sensor (3) with the sensor information and the position of the static object (5) stored in the control unit.
Need to check novelty before this filing date? Find Prior Art

Description

The invention relates to a method according to the preamble of claim 1 and to a traffic monitoring system according to the preamble of claim 10.In the case of the previously automated valve parking (AVP) type 2 systems, the infrastructure assumes responsibility for driving the vehicle, wherein the vehicle itself does not have to have any environment sensor system. For this purpose, the infrastructure builds up a control loop with the vehicle guidance via the infrastructure sensors. The trajectories are simultaneously planned centrally for all vehicles via a corresponding control unit. When installing the sensors for the infrastructure, a relatively complex calibration is necessary in order subsequently to achieve the necessary accuracy during productive operation, that is to say during the control of the vehicle or vehicles.If a sensor fails in these systems, it is damaged or its orientation is changed, recalibration must take place in order to achieve the necessary accuracy again. For this recalibration, the productive operation must be interrupted.The invention is based on the object of proposing a method for calibrating a sensor in a traffic monitoring system, which method is simple and / or highly automated and / or precise.To achieve the object, a method for calibrating a traffic monitoring system is provided. The traffic monitoring system has at least two sensors for detecting a monitoring region and a control unit connected to the sensors. At least one static object is provided in the monitoring area. The position of the static object is stored in the control unit. The method comprises the following steps: a) detecting the static object with at least one sensor and determining the position of the static object from the sensor information, and b) calibrating the sensor with the sensor information and the position of the static object stored in the control unit.A traffic monitoring system is understood to mean, in particular, systems such as, for example, the AVP type 2 system, which is suitable for automated vehicle guidance and for which no sensors are required in the vehicle. The system receives the information about the vehicle, in particular about the position and the movements of the vehicle, via the sensors installed in the traffic monitoring system and controls the vehicle from a starting point to its defined destination, for example within a parking deck, on the basis of said sensors.The monitoring area is the area that is covered or detected by a sensor and in which objects can be detected.The sensors may be any type of instrument capable of detecting an object. This can be done, for example, optically or acoustically. The optical sensors include, for example, cameras, CCD sensors or photocells. However, other methods for detecting, such as radar or LIDAR, are also conceivable.The control unit evaluates the information generated by a sensor and transmitted to the control unit, in particular position information.The sensor information serves as a basis for a wide variety of calculations of the control unit, whereby, for example, a collision of a vehicle with an object located on its route can be avoided. Sensor information can be, for example, position information, speed information, or any object information, such as, for example, turn signals, steering angle, vehicle or person, for in particular vehicles and / or persons.If the calculation reveals that a dangerous situation, for example a collision, could take place for a vehicle, the control unit can intervene, for example by slowing, stopping or calculating a different driving route for the vehicle located in the traffic monitoring space by means of corresponding control signals. A control unit is, for example, a microcontroller or central processing units (CPU). However, server-based control units are also conceivable, for example.The positions, in particular those of a static object, are stored / stored in the control unit. In addition, the control unit compares the currently detected position of an object with the position which is stored in the control unit for this object or with positions of other objects which are stored in the control unit. Thus, a collision with other objects can be avoided.The at least one static object also serves as a reference point for a sensor in order to define its position in the traffic monitoring system. Static objects are, for example, parts of the structure of the traffic monitoring space, columns, pedestrian crossings, light towers / lampers or speed or direction signs on the roadway. Furthermore, the static objects can additionally be markings provided in the traffic monitoring space, which for example enable a clear identification of the static object.The position of the static object determined with the sensor is compared with the position of the static object stored in the control unit. If the position of the static object determined from the sensor information deviates from the position stored in the control unit, a calibration of the sensor is carried out. As a result, the sensor deviating from the original calibration can be re-calibrated automatically by the control unit without the personnel being necessary for a calibration. This may also avoid time in which the system must be put into the idle state for recalibration of a sensor.Furthermore, during the calibration of the sensor, the traffic space is monitored by means of the remaining sensors. Thus, the operation of the traffic monitoring system, in particular in a parking facility, can be continued. The information of the sensor to be calibrated can be taken out of the calculations in particular during recalibration. Optionally, when setting up the traffic monitoring system, the at least one static object can be determined and / or its position can be stored during a first calibration of the at least two sensors. In particular, during the first calibration of the traffic monitoring system, a file, in particular a map of the entire traffic monitoring area, is created in which the positions of the sensors and of all static objects are stored. This file can be accessed when checking the calibration of the sensors and when recalibration of a sensor is required.In particular, the traffic space is monitored using at least three sensors. This could ensure that the traffic monitoring system can continue to operate in productive operation, since at least two further sensors are present in order to control the vehicle relatively safely to a defined destination point while the sensor deviating from its original calibration is re-calibrated.In particular, the number of sensors and the positions of the sensors are selected such that each position of the traffic monitoring area can be detected by at least two sensors, so that even in the event of a failure of one sensor, a complete detection of the traffic monitoring area takes place.Furthermore, a plurality of static objects are preferably present in the traffic monitoring system, wherein each static object can be detected by at least one sensor, in particular by at least two sensors. Preferably, a plurality of static objects can also be located in the detection range of a sensor. This allows more accurate checking and calibration of the sensors.Preferably, the traffic monitoring system can generate and transmit control commands for vehicles located on the traffic surface to the vehicles, wherein the control commands are determined depending on information determined by the sensors. The sensor information includes position and speed, for example. This allows an imminent collision with another object, for example a person who is located in the traffic space, or a point in time at which the vehicle has to turn at a branch to be determined. Depending on the prevailing situation, control commands for the brake, the throttle, the steering angle, or the flasher, for example, can be generated or changed. Thus, in particular a vehicle in a parking facility can be driven safely from a starting position to a defined target position by means of the traffic monitoring system. However, other objects which cannot be controlled by the traffic monitoring system, in particular persons, are also advantageously protected from the vehicles as a result.Furthermore, the control commands can be adjusted during the calibration of a sensor. This makes it possible to react to the omission of the sensor information during the recalibration of the one sensor, since, for example, less information could be present in this monitoring region. Thus, the vehicle would be able to be guided at reduced speed or brought to a standstill completely, for example. This additionally increases the safety of the traffic monitoring system.Preferably, the sensor is checked and calibrated automatically. This means that during the ongoing operation, the static objects are detected permanently, for example cyclically, and the calibration of the sensors is checked. Furthermore, the checking and in particular the calibration of the sensor changed from its original calibration can be started manually. This ensures that the traffic monitoring system does not automatically switch back to the productive mode, although, for example, auxiliary means required for assembly work, such as a ladder, a frame or measurement setups, would be located in the traffic space. Thus, after the work is finished, the person can manually enable and start the traffic monitoring system.To achieve the object, a traffic monitoring system is furthermore provided. The traffic monitoring system has at least two sensors for detecting a monitoring region and a control unit connected to the sensors. At least one static object is provided in the monitoring area. The position of the static object is stored in the control unit. The method comprises the following steps: a) detecting the static object with at least one sensor and determining the position of the static object from the sensor information, and b) calibrating the sensor with the sensor information and the position of the static object stored in the control unit.Preferably, the traffic space is monitored using at least three sensors.Furthermore, a plurality of static objects are present, wherein each static object can be detected by at least one sensor, in particular at least two sensors.The traffic monitoring system can preferably generate and transmit control commands for vehicles which are located on the traffic surface to the vehicles, wherein the control commands are determined as a function of information determined by the sensors.Optionally, the control commands are adjusted during calibration of a sensor.Alternatively, the calibration of the sensor changed from its original calibration can be started manually.Further advantages and features will become apparent from the following description in conjunction with the accompanying drawings. In this figure: FIG. 1 shows a schematic structure of a traffic monitoring system; FIG. 2 shows a schematic sequence of the process steps of calibrating a sensor of the traffic monitoring system by means of a static object; and FIG. 3 shows a schematic sequence as in FIG. 2 with detection of the position deviation of a sensor from the original calibration.Referring to FIG. 1, a traffic monitoring system 1 is shown. The traffic monitoring system 1 comprises two sensors 3 and a control unit, not shown here. The sensors 3 each have a monitoring area 11. Further sensors are preferably provided in the traffic space, so that the entire traffic space can be detected or monitored by sensors. In particular, the sensors are positioned such that each region of the traffic space is redundantly monitored.In addition, static objects 5, referred to below as columns 5, are present in the monitoring regions 11 of the sensors 3.If the traffic monitoring system 1 is now initialized for the first time, the sensors 3 are initially calibrated. By determining the position of the at least one column 5, the position of the sensor 3 in the traffic monitoring system is determined.FIG. 2 shows the steps of the initial calibration of a sensor 3 as part of the initialization of the traffic monitoring system 1. If the traffic monitoring system 1 is started up for the first time, the sensors 3 must be initially calibrated in order to be able to ensure an accurate production mode of the traffic monitoring system 1.The steps are listed below and carried out in this order: 1. step: detecting the static object with at least one sensor and determining the position of the static object from the sensor information; 2. step: calibrating the sensor with the sensor information and the position of the static object stored in the control unitThe steps are shortened in the drawings for reasons of clarity.After the static object 5 has been detected by the sensor 3, its position is generated as sensor information and transmitted to the control unit, not shown here. The transmission from the sensor 3 to the control unit can be effected both by cable and wirelessly.The control unit then determines the sensor position from the transmitted sensor information by the position of the static object in the monitoring region 11 of the sensor 3.For example, a map of the entire traffic space is generated and stored in the control unit, the determined positions of the static objects being stored in this map. In particular, the information of the static objects, i.e. also their position, is available for the entire traffic monitoring system.If a vehicle 7, for example by the driver thereof, is transferred to the traffic monitoring system 1 in a starting zone, the traffic monitoring system 1 establishes a connection with the vehicle 7 via wireless communication. Subsequently, a defined target 9, referred to as parking space 9 below, is defined by the control unit and the vehicle is controlled to this target 9.Control commands transmitted from the control unit to the vehicle 7 control the latter in the direction of the parking space 9 defined as the target. Here, control commands are generated in order to move the vehicle forwards or backwards accordingly or to set the corresponding steering angle of the wheels in order to control the vehicle to the parking space 9. The vehicle can implement these control commands independently, i.e. without a driver, so that the vehicle moves within the traffic area remotely by the traffic monitoring system.If a hazardous situation occurs during the movement of the vehicle 7 to the parking space 9, for example because a person is in the monitored traffic space and in the vicinity of the controlled vehicle 7, the control unit can generate and transmit control commands to the vehicle which are correspondingly adapted in order to avoid the hazardous situation. For example, the speed of the vehicle can be reduced or the vehicle can be brought completely to a standstill or a different route can be created.Should a sensor 3 be changed from its original calibration position, the effect would have on the calculation to control the vehicle 7. If a deviation between the stored position information and the transmitted position information of the sensor 3 is detected, the affected sensor 3 is recalibration.In FIG. 3, the steps as explained in FIG. 2 are extended by the step of comparing the position information of the sensor 3 with the position information stored on the control unit to form the sensor 3. This makes it possible to detect a misalignment of a sensor 3 or the need for recalibration of the sensor 3.The steps are listed below: 1. step: detecting the static object with at least one sensor and determining the position of the static object from the sensor information 2. step: comparing the position determined with the sensor with the position stored in the control unit and the calibration of the sensor is carried out if the position of the static object determined from the sensor information deviates from the position stored in the control unit. 3. step: calibrating the sensor with the sensor information and the position of the static object stored in the control unitStep 1 will not be discussed in detail, since it is identical to that of the explanation of FIG. 1.During productive operation of the traffic monitoring system 1, the sensor 3 generates position information. These can be effected both continuously and at more temporally definite intervals. If the sensor 3 is moved from its originally calibrated position, for example by a vehicle driving a mast on which the sensor 3 is mounted, the new position information is transmitted to the control unit. Subsequently, the control unit compares the new position information with the position information stored in the control unit relating to the sensor 3.If a position deviation is detected between the transmitted and the stored position information, the sensor 3 is recalibration, which recalibration can comprise, for example, automatic or manual readjustment of the sensor 3. Furthermore, correction factors can be determined for the sensor data or the sensor image, so that the positions detected with the sensor, in particular the positions of the static objects, correspond to the real positions in the traffic space.If no positional deviation is detected, step 1 is started again and a new comparison of the positional information takes place.During this time, the traffic monitoring system 1 can no longer be in productive mode, but in a sleep mode.Preferably, however, the traffic monitoring system is still in productive mode, wherein the traffic space is monitored without the sensor 3 to be calibrated. This is only used again for monitoring the traffic space when the recalibration is concluded and, in the case of a renewed check, the positions of the static objects determined with the sensor coincide with the stored positions of the static objects. In particular, the at least redundant monitoring of the traffic space by the sensors 3 always ensures complete monitoring of the traffic space. Optionally, the control commands for the vehicles 7 may be adjusted until recalibration is complete, for example by controlling the vehicle 7 through the traffic space at a reduced speed.In order to increase the accuracy of the sensors, a plurality of sensors can be combined to form a sensor group or sensor fusion.In addition, the traffic monitoring system 1 can perform a latency calibration, wherein the delay between the transmission of the control command of the vehicle 7 and the performance of the control command of the vehicle 7 can be calculated in order for the traffic monitoring system 1 to guide the vehicle 7 even more precisely, which is advantageous in narrow areas of a parking facility.List of reference characters1 Traffic monitoring system 3 Sensors / s 5 Static objects / pillars 7 Vehicle 9 Defined destination / parking space 11 Monitoring region

Claims

Method for calibrating a traffic monitoring system (1), wherein the traffic monitoring system (1) has at least two sensors (3) for detecting a monitoring region (11) and a control unit connected to the sensors (3), wherein at least one static object (5) is provided in the monitoring region (11), wherein the position of the static object (5) is stored in the control unit, having the following steps: a) detecting the static object (5) with at least one sensor (3) and determining the position of the static object (5) from the sensor information, and b) calibrating the sensor (3) with the sensor information and the position of the static object (5) stored in the control unit.Method according to Claim 1, characterized in that the position determined with the sensor (3) is compared with the position stored in the control unit, and the calibration of the sensor (3) is carried out if the position of the static object (5) determined from the sensor information deviates from the position stored in the control unit.Method according to Claim 1 or 2, characterized in that during the calibration of the sensor (3) the traffic space is monitored by means of the remaining sensors (3).Method according to one of the preceding claims, characterized in that, when setting up the traffic monitoring system (1), the at least one static object (5) can be determined and / or its position can be stored during a first calibration of the at least two sensors (3).Method according to one of the preceding claims, characterized in that the traffic space is monitored using at least three sensors (3).Method according to one of the preceding claims, characterized in that a plurality of static objects (5) are present, wherein each static object (5) can be detected by at least one sensor (3), in particular at least two sensors (3).Method according to one of the preceding claims, characterized in that the traffic monitoring system (1) can generate control commands for vehicles (7) which are located on the traffic surface and send them to the vehicles (7), wherein the control commands are determined as a function of information determined by the sensors.Method according to claim 7, characterised in that the control commands are adapted during the calibration of a sensor (3).Method according to one of the preceding claims, characterized in that the calibration of the sensor (3) which has been changed from its original calibration can be started manually.Device for calibrating a traffic monitoring system (1), wherein the traffic monitoring system (1) has at least two sensors (3) for detecting a monitoring region (11) and a control unit connected to the sensors (3), wherein at least one static object (5) is provided in the monitoring region (11), wherein the position of the static object (5) is stored in the control unit, having the following steps: a) detecting the static object (5) with at least one sensor (3) and determining the position of the static object (5) from the sensor information, b) calibrating the sensor (3) with the sensor information and the position of the static object (5) stored in the control unit.Device according to claim 10, characterised in that the traffic space is monitored by at least three sensors (3).Device according to claim 10 or 11, characterised in that a plurality of static objects (5) are present, wherein each static object (5) can be detected by at least one sensor (3), in particular at least two sensors (3).Method according to one of the preceding claims, characterized in that the traffic monitoring system can generate control commands for vehicles (7) which are located on the traffic surface and can transmit them to the vehicles (7), wherein the control commands are determined as a function of information determined by the sensors (3).Device according to claim 13, characterised in that the control commands are adapted during the calibration of a sensor (3).Device according to one of the preceding claims, characterized in that the calibration of the sensor (3) which has been changed from its original calibration can be started manually.

Citation Information

Patent Citations

  • Method and device for determining a vehicle position in a mapped environment

    DE102012016800A1

  • Method and system for self-diagnosis of at least one sensor of a system supporting an autonomously movable or movable device

    DE102018202766A1

  • Method and system for autonomous driving of a vehicle

    DE102020107108A1

  • Method for calibrating a system for infrastructure-based assistance of a motor vehicle

    DE102021214059A1