DE112022002063T5 - MODULE AND METHOD FOR ASSOCIATION OF OBJECT TRACKING DATA - Google Patents

Abstract

An object tracking data association module (100) comprising: a path data retrieval device (110) configured to provide at least one path data set of at least one path associated with at least one detected object; a sensor measurement data retrieval device (112) configured to provide at least one sensor measurement data set for the at least one path associated with the at least one detected object; and a difference parameter determination device (120) configured to provide at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object; wherein the difference parameter determination device (120) comprises an exponential function module (122) configured to perform calculations using an exponential function, and the difference parameter determination device (120) is configured to use the exponential function module (122) to calculate the at least one difference parameter using the exponential function.

Description

FIELD OF INVENTION

The invention relates to an object tracking data association module and a corresponding method for associating object tracking data.

BACKGROUND

Object detection and object tracking can have a wide range of applications in different fields, for example in driving, more specifically in motor vehicles on the road. Object detection and object tracking complement each other. Although object detection may involve the detection of an object in an image, it would not provide information about the object's motion history. Therefore, object tracking can complement object detection by taking into account a motion history of an object. As a result, object tracking can be useful for tracking an object for a period of time, for example in a video. However, object tracking can be complicated when multiple objects are to be tracked, since a currently detected object may need to be associated with a previously detected object, for example because the previously detected object may have moved to a new position, a previously detected object may no longer be detected by a sensor because it has, for example, left a field of view of the camera, or a newly detected object may currently be detected by a sensor because it has entered the field of view of the camera.

BRIEF DESCRIPTION

One object is to provide an object tracking data association module that is effective in performing data association for object tracking or an effective method for associating object tracking data.

According to a first aspect of the invention, there is provided an object tracking data association module comprising: a path data retrieval device configured to provide at least one path data set of at least one path associated with at least one detected object; a sensor measurement data retrieval device configured to provide at least one sensor measurement data set for the at least one path associated with the at least one detected object; and a difference parameter determination device configured to determine at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object; wherein the difference parameter determination device comprises an exponential function module configured to perform calculations using an exponential function, and the difference parameter determination device is configured to use the exponential function module to calculate the at least one difference parameter using the exponential function.

The object tracking data association module comprises the difference parameter determination device configured to determine the at least one difference parameter using the exponential function, which would advantageously enable data association to be performed effectively.

Optionally, the difference parameter determination device comprises a correlation parameter determination device which is designed to determine at least one correlation parameter which defines a correlation between the at least one sensor measurement data set and the at least one path data set.

The correlation parameter determination device is designed to determine the at least one correlation parameter that defines a correlation between the at least one sensor measurement data set and the at least one path data set, which would advantageously enable consideration of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the difference parameter determination device comprises a statistical distance calculator which is designed to statistically calculate the at least one correlation parameter.

The statistical distance calculator is configured to statistically calculate the at least one correlation parameter, which would advantageously enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the exponential function has an exponent that has the at least one correlation parameter.

The difference parameter determination device is designed to determine the at least one diff ference parameter using the exponential function having the exponent having the at least one correlation parameter, which would advantageously enable data association to be performed effectively.

Optionally, the exponent is a negative multiple of the at least one correlation parameter.

The difference parameter determining means is configured to determine the at least one difference parameter using the exponential function having the exponent that is the negative multiple of the at least one correlation parameter, which would advantageously enable data association to be performed effectively.

Optionally, the difference parameter determination device comprises a Mahalanobis distance calculator configured to calculate the at least one correlation parameter using a Mahalanobis distance formula, wherein the at least one correlation parameter comprises a Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

The correlation parameter determination device comprises the Mahalanobis distance calculator, which is configured to calculate the at least one correlation parameter using the Mahalanobis distance formula, which would advantageously enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the exponential function has an exponent that represents the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

The difference parameter determination device is configured to determine the at least one difference parameter using the exponential function having the exponent representing the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set, which would advantageously enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the exponent has a negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

The difference parameter determination device is configured to determine the at least one difference parameter using the exponential function having the exponent that is the negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set, which would advantageously enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the difference parameter determination device is designed to calculate the at least one difference parameter by subtracting a value of the exponential function from one.

The difference parameter determination device is designed to determine the at least one difference parameter by subtracting a value of the exponential function from one, which would advantageously enable data association to be carried out effectively.

The object tracking data association module may further comprise an existence probability determiner configured to accept the at least one difference parameter as input and output at least one existence probability parameter for the at least one path associated with the at least one detected object.

The existence probability determination device is designed to output the at least one existence probability parameter, which would advantageously enable an effective implementation of data association.

Optionally, the existence probability determination device comprises a machine learning system configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object.

The machine learning system is configured to output the at least one existence probability parameter, which would advantageously enable effective performance of data association.

Optionally, the existence probability determination device comprises an LSTM network module configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object.

The LSTM network module is configured to output the at least one existence probability parameter, which would advantageously enable data association to be performed effectively. Furthermore, the LSTM network module may advantageously be trained to enable the object tracking data association module to perform end-to-end object tracking with back propagation together with an estimator, e.g. a Kalman filter.

Optionally, the path data retrieval device is configured to provide the at least one path data set comprising an active path data set of an active path associated with a previously detected object.

The path data retrieval device is configured to provide the at least one path data set comprising the active path data set of the active path, which would advantageously enable consideration of data relating to the previously detected object contained in the active path data set.

Optionally, the path data retrieval device is configured to provide the at least one path data set comprising an inactive path data set of an inactive path associated with a newly detected object.

The path data retrieval device is configured to provide the at least one path data set comprising the inactive path data set, which would advantageously enable consideration of data relating to the newly detected object.

The object tracking data association module may further comprise a threshold comparator configured to compare the at least one existence probability parameter of the at least one path with an existence threshold.

The object tracking data association module may further comprise a decision device configured to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold.

The decision device is designed to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold, which would advantageously enable an effective association of a currently detected object with the active path.

The object tracking data association module may further comprise a decision device configured to terminate the active path if the at least one existence probability parameter of the active path is below the existence threshold.

The decision device is configured to terminate the active path if the at least one existence probability parameter of the active path is below the existence threshold, which would advantageously effectively prevent the object tracking data association module from tracking the previously detected object that is no longer detected by a sensor because it has, for example, left the field of view of the camera.

The object tracking data association module may further comprise a decision device configured to initialize the inactive path to another active path if the at least one existence probability parameter of the inactive path is above the existence threshold.

The decision device is designed to initialize the inactive path to the other active path if the at least one existence probability parameter of the inactive path is above the existence threshold, which would advantageously enable effective tracking of the newly detected object.

The object tracking data association module may further comprise a sensor reliability value retriever configured to obtain at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set for the at least one path associated with the at least one detected object.

The sensor reliability value retrieval device is designed to obtain at least one sensor reliability parameter, which advantageously allows consideration of the at least one sensor reliability parameter.

The object tracking data association module may further comprise a threshold comparator configured to compare the at least one sensor reliability parameter to a sensor reliability threshold.

The object tracking data association module may further comprise a decision device configured to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold.

The decision device is configured to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold, which would advantageously enable identification and ignoring of unreliable sensor data.

An object tracking module may include the object tracking data association module.

A motor vehicle may include the object tracking data association module.

Any features or steps disclosed in the context of the first aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally in the inventions, where possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the first aspect of the invention and generally in the inventions, where possible.

According to a second aspect of the invention, an object tracking data association module is provided, comprising: a path data retrieval device configured to provide at least one path data set of at least one path associated with at least one detected object; wherein the path data retrieval device is configured to provide the at least one path data set comprising an active path data set of an active path associated with a previously detected object; wherein the path data retrieval device is configured to provide the at least one path data set comprising an inactive path data set of an inactive path associated with a newly detected object; a sensor measurement data retrieval device configured to provide at least one sensor measurement data set for the at least one path associated with the at least one detected object; a difference parameter determination device configured to determine at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object; wherein the difference parameter determination device comprises an exponential function module configured to perform calculations using an exponential function, and the difference parameter determination device is configured to use the exponential function module to calculate the at least one difference parameter using the exponential function; wherein the difference parameter determination device comprises a correlation parameter determination device configured to determine at least one correlation parameter that defines a correlation between the at least one sensor measurement data set and the at least one path data set; wherein the difference parameter determination device comprises a statistical distance calculator configured to statistically calculate the at least one correlation parameter; wherein the difference parameter determination device comprises a Mahalanobis distance calculator configured to calculate the at least one correlation parameter using a Mahalanobis distance formula, wherein the at least one correlation parameter comprises a Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; wherein the difference parameter determination device is designed to calculate the at least one difference parameter by subtracting a value of the exponential function from one; wherein the exponential function has an exponent which has the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; wherein the exponent has a negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; an existence probability determination device which is designed to accept the at least one difference parameter as input and to output at least one existence probability parameter for the at least one path associated with the at least one detected object; wherein the existence probability determination device comprises a machine learning system configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object; wherein the existence probability determination device comprises an LSTM network module configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object; a threshold comparator configured to compare the at least one existence probability parameter of the at least one path with an existence threshold; a decision device configured to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold; wherein the decision device is configured to terminate an active path if the at least one existence probability parameter of the active path is below the existence threshold; wherein the decision device is configured to initialize the inactive path to an active path if the at least one existence probability parameter of the inactive path is above the existence threshold; and a sensor reliability value retrieval device configured to obtain at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set for the at least one path associated with the at least one detected object; wherein the threshold comparator is configured to compare the at least one sensor reliability parameter with a sensor reliability threshold; wherein the decision device is configured to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold.

The object tracking data association module would advantageously enable data association to be performed effectively. The correlation parameter determination device is configured to determine the at least one correlation parameter defining a correlation between the at least one sensor measurement data set and the at least one path data set, which would advantageously enable consideration of a correlation between the at least one sensor measurement data set and the at least one path data set. The statistical distance calculator is configured to statistically calculate the at least one correlation parameter, which would advantageously enable reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set. The correlation parameter determination device comprises the Mahalanobis distance calculator, which is configured to calculate the at least one correlation parameter using the Mahalanobis distance formula, which would advantageously enable reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set. The LSTM network module can advantageously be trained to enable the object tracking data association module, together with an estimator, for example a Kalman filter, to perform end-to-end object tracking with back propagation. The path data retrieval device is configured to provide the at least one path data set comprising the active path data set of the active path, which would advantageously enable consideration of data relating to the previously detected object contained in the active path data set. The path data retrieval device is configured to provide the at least one path data set comprising the inactive path data set, which would advantageously enable consideration of data relating to a newly detected object. The decision device is configured to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold, which would advantageously enable effective association of a just detected object with the active path. The decision-making device is designed to terminate the active path if the at least one existence probability parameter of the active path is below the existence threshold, which would advantageously effectively prevent the object tracking data association module from tracking the previously detected object that is no longer detected by any sensor because it has, for example, left the field of view of the camera. The decision-making device is designed to initialize the inactive path to the other active path if the at least one existence probability parameter of the inactive path is above the existence threshold, which would advantageously enable effective tracking of the newly detected object. The sensor reliability value retrieval device is designed to obtain at least one sensor reliability parameter, which would advantageously allow consideration of the at least one sensor reliability parameters. The decision device is designed to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold, which would advantageously enable identification and ignoring of unreliable sensor data.

Any features or steps disclosed in the context of the second aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally in the inventions, where possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the second aspect of the invention and generally in the inventions, where possible.

According to a third aspect of the invention, there is provided a method for associating object tracking data, comprising the acts of: providing, by a processor, at least one path record of at least one path associated with at least one detected object; providing, by the processor, at least one sensor measurement record for the at least one path associated with the at least one detected object; and determining, by the processor, at least one difference parameter defining a relationship between the at least one sensor measurement record and the at least one path record of the at least one path associated with the at least one detected object using an exponential function.

The method for associating object tracking data determines the at least one difference parameter using the exponential function, which would advantageously enable data association to be performed effectively.

The method for associating object tracking data may further comprise the act of determining, by the processor, at least one correlation parameter defining a correlation between the at least one sensor measurement data set and the at least one path data set.

Advantageously, this would allow consideration of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the act of determining the at least one correlation parameter comprises the act of statistically calculating, by the processor, the at least one correlation parameter.

Advantageously, this would enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the act of determining the at least one difference parameter comprises the act of determining, by the processor, the at least one difference parameter using the exponential function having an exponent comprising the at least one correlation parameter.

Advantageously, this would enable effective implementation of data association.

Optionally, the act of determining the at least one difference parameter comprises the act of determining, by the processor, the at least one difference parameter using the exponential function having the exponent that is a negative multiple of the at least one correlation parameter.

Advantageously, this would enable effective implementation of data association.

Optionally, the act of determining the at least one difference parameter comprises the act of calculating, by the processor, the at least one correlation parameter using a Mahalanobis distance formula, wherein the at least one correlation parameter comprises a Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

Advantageously, this would enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the act of determining the at least one difference parameter comprises the act of determining, by the processor, the at least one difference parameter using the exponential function having an exponent that represents the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

Advantageously, this would enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the act of determining the at least one difference parameter comprises the act of determining, by the processor, the at least one difference parameter using the exponential function having the exponent that is a negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set.

Advantageously, this would enable a reliable and appropriate analysis of a correlation between the at least one sensor measurement data set and the at least one path data set.

Optionally, the act of determining the at least one difference parameter comprises the act of subtracting, by the processor, a value of the exponential function from one.

Advantageously, this would enable effective implementation of data association.

The method for associating object tracking data may further comprise the act of providing, by the processor, the at least one difference parameter as input to an existence probability determination device.

Optionally, the act of providing the at least one difference parameter as input to the existence probability determination device may comprise the act of providing the at least one difference parameter by the processor as input to a machine learning system.

Optionally, the act of providing the at least one difference parameter as input to the existence probability determiner may comprise the act of providing the at least one difference parameter by the processor as input to an LSTM network module.

The method for associating object tracking data may further comprise the act of determining, by the processor, at least one existence probability parameter for the at least one path associated with the at least one detected object using the existence probability determining means.

Advantageously, this would enable effective implementation of data association.

Optionally, the act of determining the at least one existence probability parameter using the existence probability determining device comprises the act of determining, by the processor, the at least one existence probability parameter for the at least one path associated with the at least one detected object using the machine learning system.

Advantageously, this would enable effective implementation of data association.

Optionally, the act of determining the at least one existence probability parameter using the existence probability determining device comprises the act of determining, by the processor, the at least one existence probability parameter for the at least one path associated with the at least one detected object using the LSTM network module.

Advantageously, this would allow data association to be performed effectively. Furthermore, the LSTM network module may advantageously be trained to allow the object tracking data association module to perform end-to-end object tracking with back propagation together with an estimator, e.g. a Kalman filter.

Optionally, the act of providing the at least one path record comprises the act of providing, by the processor, an active path record of an active path associated with a previously detected object.

Advantageously, this would allow consideration of data related to the previously detected object contained in the active path dataset.

Optionally, the act of providing the at least one path record comprises the act of providing, by the processor, the at least one path record comprising an inactive path record of an inactive path associated with a newly recognized object.

Advantageously, this would allow consideration of data relating to the newly detected object.

The method for associating object tracking data may further comprise the act of comparing, by the processor, the at least one existence probability parameter of the at least one path to an existence threshold.

The method for associating object tracking data may further comprise the act of performing an association for the active path, by the processor, when the at least one existence probability parameter of the active path is above the existence threshold.

Advantageously, this would enable an effective association of a currently detected object with the active path.

The method for associating object tracking data may further comprise the act of terminating, by the processor, an active path when the at least one existence probability parameter of the active path is below the existence threshold.

Advantageously, this would effectively stop tracking the previously detected object that is no longer detected by a sensor because, for example, it has left the camera's field of view.

The method for associating object tracking data may further comprise the act of initializing, by the processor, the inactive path to an active path when the at least one existence probability parameter of the inactive path is above the existence threshold.

Advantageously, this would enable effective tracking of the newly detected object.

The method for associating object tracking data may further comprise an act of obtaining, by the processor, at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set for the at least one path associated with the at least one detected object.

Advantageously, this would allow consideration of at least one sensor reliability parameter.

The method for associating object tracking data may further comprise an act of comparing, by the processor, the at least one sensor reliability parameter to a sensor reliability threshold.

The method for associating object tracking data may further comprise an act of ignoring, by the processor, the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold.

Advantageously, this would enable identification and ignoring of unreliable sensor data.

Any features or steps disclosed in the context of the third aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally in the inventions, where possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the third aspect of the invention and generally in the inventions, where possible.

According to a fourth aspect of the invention, there is provided a computer-implemented method for associating object tracking data, comprising performing the following actions on a processor: providing at least one path data set of at least one path associated with at least one detected object; providing at least one sensor measurement data set for the at least one path associated with the at least one detected object; and determining at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object using an exponential function.

The method for associating object tracking data determines the at least one difference parameter using the exponential function, which would advantageously enable data association to be performed effectively.

All features or steps disclosed in the context of the fourth aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally be used in the inventions, wherever possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the fourth aspect of the invention and generally in the inventions, wherever possible.

According to a fifth aspect of the invention, there is provided a non-transitory computer readable medium having instructions stored thereon that, when executed on a processor, perform a method for associating object tracking data comprising the acts of: providing at least one path record of at least one path associated with at least one detected object; providing at least one sensor measurement record for the at least one path associated with the at least one detected object; and determining at least one difference parameter defining a relationship between the at least one sensor measurement record and the at least one path record of the at least one path associated with the at least one detected object using an exponential function.

The method for associating object tracking data determines the at least one difference parameter using the exponential function, which would advantageously enable data association to be performed effectively.

Any features or steps disclosed in the context of the fifth aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally in the inventions, where possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the fifth aspect of the invention and generally in the inventions, where possible.

According to a sixth aspect of the invention, there is provided a method for associating object tracking data, comprising the acts of: providing at least one path data set of at least one path associated with at least one detected object; providing at least one sensor measurement data set for the at least one path associated with the at least one detected object; and determining at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object using an exponential function.

The method for associating object tracking data determines the at least one difference parameter using the exponential function, which would advantageously enable data association to be performed effectively.

Any features or steps disclosed in the context of the sixth aspect of the invention may also be used in combination with and/or in the context of other aspects of the invention and generally in the inventions, where possible. Furthermore, any features or steps disclosed in the context of another aspect of the invention may also be used in combination with and/or in the context of the sixth aspect of the invention and generally in the inventions, where possible.

In this summary, in the description below, in the claims that follow, and in the accompanying drawings, reference is made to specific features (including method steps) of the invention. It is to be understood that the disclosure of the invention in this specification includes all possible combinations of such specific features. For example, where a specific feature is disclosed in the context of a specific aspect or embodiment of the invention or a specific claim, that feature may also be used in combination with and/or in the context of other specific aspects and embodiments of the invention and in the inventions generally, where possible.

In this summary, in the description below, in the claims that follow, and in the accompanying drawings, reference is made herein to a method comprising two or more defined steps, wherein the defined steps may be performed in any order or simultaneously (except where the context precludes this possibility), and the method may comprise one or more further steps performed before any of the defined steps, between two of the defined steps, or after all of the defined steps (except where the context precludes this possibility).

As used in this summary, the following description, the following claims, and the accompanying drawings, the term “comprising” and grammatical equivalents thereof are used to mean used to mean that other components, ingredients, steps, etc. are optionally present. For example, an article "having" (or "the") components A, B, and C ("having") may consist of (that is, contain only) components A, B, and C, or it may contain not only components A, B, and C but also one or more other components.

As used in this summary, the description below, the claims below, and the accompanying drawings, the term "at least" followed by a number is used to denote the beginning of a range that begins with that number (and which, depending on the variable being defined, may be a range with an upper limit or no upper limit). For example, "at least 1" means 1 or more than 1. The term "at most" followed by a number is used to denote the end of a range that ends with that number (and which, depending on the variable being defined, may be a range with 1 or 0 as its lower limit or a range with no lower limit). For example, "at most 4" means 4 or less than 4, and "at most 40%" means 40% or less than 40%. In this specification, when a range is given as "(first number) to (second number)" or "(first number) - (second number)", it refers to a range whose lower limit is the first number and whose upper limit is the second number. For example, 25 to 100 mm refers to a range whose lower limit is 25 mm and whose upper limit is 100 mm.

As used in this summary, the description below, the claims below, and the accompanying drawings, the term "volatile memory" refers to any type of computer memory in which the contents of the memory are lost when power is not supplied to the computer. Random access memory (RAM) is an example of such a type of volatile memory. As used in the summary above, the description, the claims below, and the accompanying drawings, the term "non-volatile memory" or the term "non-transitory computer-readable medium" refers to any type of computer memory in which the contents of the memory are preserved even when power is not supplied to the computer. Hard disk drives and solid state drives (SSDs) are examples of types of non-volatile memory or non-transitory medium.

As used in this abstract, the description below, the claims that follow, and the accompanying drawings, the term "machine learning system" refers to a computer system capable of learning without direct programming instructions. A machine learning system applies statistical modeling based on data input and without direct programming instructions to recognize patterns and improve performance. A machine learning system builds a statistical model through a training or learning process that involves inputting data into the machine learning system. The four basic categories of learning processes are supervised learning using labeled data sets, unsupervised learning using unlabeled data sets, semi-supervised learning using a mix of labeled data sets and unlabeled data sets, and reinforcement learning which involves learning by trial and error. Decision tree, support vector machine, and neural network are examples of types of machine learning systems.

As used in this abstract, the description below, the claims that follow, and the accompanying drawings, the term "neural network" or the term artificial neural network refers to a type of machine learning algorithm that uses a network of nodes, edges, and layers. The first layer of a neural network includes input nodes that accept data inputs from a data set. The input nodes then send information about the edges to the nodes in the next layer. Each edge has an activation function that is changeable during a training process. The final layer of the neural network has the output nodes that provide data outputs of the neural network. During the training process, the data outputs of the neural network are compared with the actual outputs of the data set. The differences between the data outputs of the neural network and the actual outputs of the data set are measured and called the error value. The error value is then fed back into the neural network, which changes its activation functions to minimize the error value. The training process is an iterative process. After training the neural network, the trained neural network can then be used to predict a data output from a specific data input. An LSTM network is an example of a type of artificial neural network.

As set forth in this summary, the following description, the following claims, and the accompanying drawings As used herein, the term “object being detected” or “detected object” refers to an object that was detected, for example, by a sensor, such as an ultrasonic sensor, object and distance sensors, or a camera, at a specific time, for example in a specific frame of a video. Consequently, a currently detected object would refer to an object that was detected at the specific time, for example in the specific frame of the video. Accordingly, a previously detected object would refer to an object that was detected before the specific time, for example in the previous frame of the video. Therefore, a newly detected object would refer to an object that was just detected at the specific time and not before the specific time, because the newly detected object has, for example, just entered the field of view of a camera.

As used in this summary, the description below, the claims below, and the accompanying drawings, the term "path" refers to a placeholder included in an object tracking data association module that can be used to associate with a detected object. As a result, data associated with a detected object can be associated with a path. Thus, not only can data associated with a detected object that is currently obtained be associated with a path at a specific time, but previously obtained data associated with the detected object or predicted data associated with the detected object can also be associated with the path. An active path would be associated with a previously detected object, and an inactive path would be associated with either an undetected object or a newly detected object.

As used in this summary, the description below, the claims that follow, and the accompanying drawings, the term "processor" refers to a computer component configured to perform computations, make decisions, execute instructions, process data, or control other computer components. A central processing unit (CPU) and a graphics processing unit (GPU) are examples of types of processors.

BRIEF DESCRIPTION OF THE DRAWINGS

• 1 ein Objektverfolgungsmodul, das funktionell mit einem Sensormodul verbunden ist;
• 2 ein Objektverfolgungsdatenassoziationsmodul;
• 3 eine Differenzparametertabelle, die Werte von Differenzparametern aufweist;
• 4 eine Existenzwahrscheinlichkeitstabelle, die Werte von Existenzwahrscheinlichkeitsparametern aufweist;
• 5 ein Kraftfahrzeug, das das Objektverfolgungsdatenassoziationsmodul von 2 aufweist; und
• 6 ein Diagramm für ein Verfahren zur Assoziation von Objektverfolgungsdaten.

These and other features, aspects and advantages will be better understood by reference to the following description, the appended claims and the accompanying drawings, in which:

• 1 an object tracking module operatively connected to a sensor module;
• 2 an object tracking data association module;
• 3 a difference parameter table containing values of difference parameters;
• 4 an existence probability table containing values of existence probability parameters;
• 5 a motor vehicle that uses the object tracking data association module of 2 and
• 6 a diagram for a method for associating object tracking data.

In the drawings, like parts are identified by like reference numerals.

DESCRIPTION

In the foregoing summary, in this description, in the following claims, and in the accompanying drawings, reference is made to specific features (including method steps) of the invention. It is to be understood that the disclosure of the invention in this specification includes all possible combinations of such specific features. For example, where a specific feature is disclosed in the context of a specific aspect or embodiment of the invention or a specific claim, that feature may also be used in combination with and/or in the context of other specific aspects and embodiments of the invention and in the inventions generally, where possible.

1 shows an object tracking module 102 that is operatively connected to a sensor module 104. The object tracking module 102 may include an object tracking data association module 100 and an estimator, such as a Kalman filter 106, optionally connected in a feedback loop. The sensor module 104 may include at least one sensor, such as an ultrasonic sensor, object and distance sensors, or a camera. The sensor module 104 is configured to receive sensor measurement data and to provide the sensor measurement data to the object tracking module 102.

2 shows the object tracking data association module 100, which includes a path data retrieval device 110, a sensor measurement data retrieval device 112, a sensor reliability value Retrieval device 114, a difference parameter determination device 120, an existence probability determination device 130, a threshold comparator 140 and a decision device 142.

The sensor measurement data retriever 112 is configured to obtain sensor measurement data associated with at least one object detected by the sensor module 104. Sensor measurement data associated with each object detected at a specific time, for example in the specific frame of the video, may be included in a respective sensor measurement data set. A sensor measurement data set may include at least one of position data, velocity data, or acceleration data. A sensor measurement data set would also be associated with a path. As a result, each path associated with a respective detected object may include more than one measurement data set, since more than one sensor may have detected the same object. Therefore, the sensor measurement data retriever 112 is configured to provide at least one sensor measurement data set for at least one path associated with at least one detected object.

The path data retrieval device 110 is configured to provide at least one path data set of at least one path associated with at least one detected object. Each path data set may include at least one of position data, velocity data, or acceleration data associated with the respective detected object. A path data set may include either an active path data set associated with an active path or an inactive path data set associated with an inactive path. The path data retrieval device 110 is configured to provide at least one path data set including an active path data set of an active path associated with a previously detected object. The path data retrieval device 110 is also configured to provide at least one path data set including an inactive path data set of an inactive path associated with a newly detected object.

The sensor reliability value retriever 114 is configured to obtain a respective sensor reliability parameter associated with each sensor used to provide sensor measurement data. For example, a sensor may have a sensor reliability parameter provided by its manufacturer or obtained from tests performed on the sensor to determine its detection reliability. As a result, the sensor reliability value retriever 114 is configured to obtain at least one sensor reliability parameter associated with at least one sensor used to provide at least one sensor measurement data set for at least one path associated with at least one detected object.

The difference parameter determination device 120 has an exponential function module 122 and a correlation parameter determination device 124 which has a statistical distance calculator 126 which has a Mahalanobis distance calculator 128.

The difference parameter determination device 120 is configured to determine at least one difference parameter that defines a relationship between at least one sensor measurement data set and at least one path data set of at least one path associated with at least one detected object.

A difference parameter defines a relationship between each sensor measurement data set and a respective path data set of a corresponding path associated with a detected object. A difference parameter may be calculated using a difference parameter equation. A difference parameter equation may include an exponential function having an exponent that may include a correlation parameter that defines a correlation between each sensor measurement data set and a respective path data set. A difference parameter equation may also include an exponential function having an exponent that may include a negative multiple of a correlation parameter. A correlation parameter may be calculated statistically. A correlation parameter may be calculated using a Mahalanobis distance formula such that the correlation parameter includes a Mahalanobis distance between each sensor measurement data set and a respective path data set. A difference parameter equation may include an exponential function having an exponent that may include a Mahalanobis distance between each sensor measurement data set and a respective path data set. A difference parameter equation may also include an exponential function having an exponent that may be a negative multiple of a Mahalanobis distance between each sensor measurement data set and a respective path data set. A difference parameter may be calculated by subtracting a value of the exponential function from one.

The exponential function module 122 is for performing calculations using of exponential functions. The difference parameter determination device 120 is configured to use the exponential function module 122 to calculate a difference parameter using an exponential function. The correlation parameter equation 124 is configured to determine a correlation parameter that defines a correlation between each sensor measurement data set and a respective path data set. The statistical distance calculator 126 is configured to statistically calculate a correlation parameter. The Mahalanobis distance calculator 128 is configured to calculate a correlation parameter using a Mahalanobis distance formula, wherein each correlation parameter has a Mahalanobis distance between a respective sensor measurement data set and a corresponding path data set.

wobei m ein Messdatenvektor ist, der Daten aus einem Sensormessdatensatz aufweist; t ein Pfaddatenvektor ist, der Daten aus einem Pfaddatensatz aufweist; und S eine Kovarianzmatrix von m und t ist. Ein unter Verwendung einer Mahalanobis-Abstandsformel berechneter CP wäre ein positive Zahl.A correlation parameter, CP, can be calculated using a Mahalanobis distance formula as follows: $$CP=\sqrt{{\left(\overrightarrow{m}-\overrightarrow{t}\right)}^T{S}^{-1}\left(\overrightarrow{m}-\overrightarrow{t}\right)}$$

where m is a measurement data vector comprising data from a sensor measurement data set; t is a path data vector comprising data from a path data set; and S is a covariance matrix of m and t. A CP calculated using a Mahalanobis distance formula would be a positive number.

wobei a eine positive Zahl, beispielsweise die Eulersche Zahl, ist; und CP ein Korrelationsparameter ist.A difference parameter equation can be as follows: $$DP=1-a^{-CP}$$

where a is a positive number, for example Euler's number; and CP is a correlation parameter.

-CP is an example of an exponent suitable for calculating a difference parameter, and thus a^{-CP} is an example of an exponential function suitable for calculating a difference parameter. In addition, CP can be calculated statistically using a Mahalanobis distance formula. Since CP is a positive number, -CP would be a negative number.

If there were only one sensor and only one detected object, then there would be only one correlation parameter and accordingly only one difference parameter. If there were more than one sensor and/or more than one detected object, then there would be more than one correlation parameter and accordingly more than one difference parameter. In this case, values of the difference parameters can be stored in a difference parameter matrix or a difference parameter table 150, as in 3 presented, provided.

A difference parameter table 150 includes values of difference parameters. A difference parameter table 150 may have as many rows and as many columns as desired and practical. The first row from the top of a difference parameter table 150 includes path existence values of paths. Each path has a respective path existence value. If a path existence value of a path is above a path existence threshold, the path would be an active path. If a path existence value of a path is below a path existence threshold, the path would be an inactive path. The first column from the left of a difference parameter table 150 includes values of sensor reliability parameters associated with sensors used to provide sensor measurement data of paths associated with detected objects. Each row except the first row from the top of a difference parameter table 150 represents a measurement attempt by a sensor at a specific time. Further, a sensor may make more than one measurement attempt at a specific time. Additionally, more than one object may be detected in a measurement attempt. As a result, the first cell from the left of each row would have a value of a sensor reliability parameter of a sensor making a measurement attempt represented by that row. Therefore, the top left cell of a difference parameter table 150 would be empty, while all other cells except cells in the first row from the top and the first column from the left would have values of difference parameters.

The existence probability determiner 130 may include a machine learning system 132, which may include an LSTM network module 134.

The existence probability determination device 130 is configured to accept at least one difference parameter as input and to output at least one existence probability parameter for at least one path associated with at least one detected object. The existence probability determination device 130 is configured to output the same number of existence probability parameters as the number of difference parameters provided as input to the existence probability determination device 130. Thus, if there were more than one difference parameter for each path associated with a respective detected object, the existence probability determination device 130 would accept the more than one difference parameter as input and output a corresponding sufficient number of existence probability parameters.

The machine learning system 132 is configured to accept at least one difference parameter as input and output at least one existence probability parameter for at least one path associated with at least one detected object. The machine learning system 132 is configured to output the same number of existence probability parameters as the number of difference parameters provided as input to the machine learning system 132. Thus, if there were more than one difference parameter for each path associated with a respective detected object, the machine learning system 132 would accept the more than one difference parameter as input and output a corresponding number of existence probability parameters.

The LSTM network module 134 is configured to accept at least one difference parameter as input and output at least one existence probability parameter for at least one path associated with at least one detected object. The LSTM network module 134 is configured to output the same number of existence probability parameters as the number of difference parameters provided as input to the LSTM network module 134. Thus, if there were more than one difference parameter for each path associated with a respective detected object, the LSTM network module 134 would accept the more than one difference parameter as input and output a corresponding number of existence probability parameters. The LSTM network module may advantageously be trained to enable the object tracking data association module, together with the estimator, e.g. the Kalman filter 106, to perform end-to-end object tracking with back propagation.

Values in a difference parameter table 150 may be provided to the LSTM network module 134, and values of existence probability parameters obtained may then be entered into an existence probability matrix or an existence probability table 152, as in 4 shown, entered.

An existence probability table 152 includes values of existence probability parameters. An existence probability table 152 may have as many rows and as many columns as a corresponding difference parameter table 150. The first row from the top of an existence probability table 152 includes path existence values of paths. Each path has a respective path existence value. If a path existence value of a path is above a path existence threshold, the path would be an active path. If a path existence value of a path is below a path existence threshold, the path would be an inactive path. The first column from the left of an existence probability table 152 includes values of sensor reliability parameters associated with sensors used to provide sensor measurement data of paths associated with detected objects. Each row except the first row from the top of an existence probability table 152 represents a measurement attempt by a sensor at a specific time. Furthermore, the top left cell of an existence probability table 152 would be empty, while all other cells except cells in the first row from the top and the first column from the left would have values of existence probability parameters.

The threshold comparator 140 is configured to compare at least one existence probability parameter of a path to an existence threshold. The existence threshold may be any suitable value, such as 0.5. The threshold comparator 140 is also configured to compare a sensor reliability parameter to a sensor reliability threshold. The sensor reliability threshold may be any suitable value, such as 0.5. The threshold comparator 140 may also be configured to compare a difference parameter to a difference threshold. The difference threshold may be any suitable value, such as 0.5.

The decision device 142 is designed to perform an association for an active path if an existence probability parameter of the active path is above the existence threshold (for example, a path associated with column 3 and a path associated with column 6 of the existence probability table 152 of 4 is associated). The decision device 142 is also designed to terminate an active path if an existence probability parameter of the active path is below an existence threshold. The decision device 142 may also be designed to terminate an active path if all existence probability parameters of the active path are below an existence threshold (for example, a path associated with column 7 of the existence probability table 152 of 4 Furthermore, the decision device 142 is designed to convert an inactive path to an active path to initialize if at least one existence probability parameter of the inactive path is above an existence threshold (for example, a path that matches column 2 of the existence probability table 152 of 4 associated with).

5 shows a motor vehicle 108 having the object tracking data association module 100.

6 shows a diagram for a method 200 for associating object tracking data using the object tracking data association module 100. The acts or steps of the method 200 for associating object tracking data may be performed by a processor, such as a central processing unit of a computer.

At step 202, the method 200 initializes for associating object tracking data. At step 204, the sensor module 104 detects at least one object and obtains at least one sensor measurement data set for each detected object. At step 206, the at least one sensor measurement data set is provided for each path associated with a respective detected object. At step 208, at least one path data set is provided, wherein each path data set of a respective path is associated with a corresponding detected object. The at least one path data set may include at least one active path data set, wherein each active path data set of a respective active path is associated with a corresponding previously detected object. The at least one path data set may include at least one inactive path data set, wherein each inactive path data set of a respective inactive path is associated with a corresponding newly detected object.

At step 210, at least one difference parameter is determined, each difference parameter defining a relationship between a respective sensor measurement data set and a corresponding path data set. The process of step 210 begins at step 212. At step 214, at least one correlation parameter is determined, each correlation parameter defining a correlation between a respective sensor measurement data set and a corresponding path data set. At step 216, the process of step 214 begins. At step 218, each correlation parameter is calculated using a Mahalanobis distance formula such that each correlation parameter has a Mahalanobis distance between a respective sensor measurement data set and a corresponding path data set. The process of step 214 ends at step 220.

Then, at step 222, a respective value of an exponential function having an exponent that is a negative multiple of each correlation parameter is calculated. At step 224, each difference parameter is calculated by subtracting a value of the exponential function from one. The process of step 210 ends at step 226.

At step 228, at least one existence probability parameter is determined. At step 230, the process of step 228 begins. At step 232, the at least one difference parameter is provided to the machine learning system 132, such as the LTSM network module 134. The machine learning system 132 or the LTSM network module 134 would have been trained to accept the at least one difference parameter as input and output a respective existence probability parameter for each difference parameter provided. As a result, the machine learning system 132 or the LTSM network module 134 outputs the at least one existence probability parameter at step 234. The process of step 228 ends at step 236.

At step 238, at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set is obtained. At step 240, the at least one sensor reliability parameter is compared to a sensor reliability threshold. At step 242, the at least one existence probability parameter is compared to an existence threshold.

At step 244, an association is performed for an active path if an existence probability parameter of the active path is below the existence threshold. As a result, assuming that the existence threshold is 0.5, an existence probability parameter in row 3, column 3 of the existence probability table of 4 linked to a path (for example, path 2) that corresponds to column 3 of the existence probability table of 4 is associated because a value of the existence probability parameter in row 3, column 3 is above 0.5.

At step 246, an active path is terminated if an existence probability parameter of the active path is below the existence threshold. As a result, assuming that the existence threshold is 0.5, a path (for example, path 6) matching column 7 of the existence probability table of 4 associated, terminated, since none of the existence probabilities parameter of the path (for example path 6) has a value greater than 0.5.

At step 248, an inactive path is initialized to an active path if an existence probability parameter of the inactive path is above the existence threshold. As a result, assuming that the existence threshold is 0.5, a path (for example, path 1) matching column 2 of the existence probability table of 4 is initialized because a value of the existence probability parameter in row 2, column 2 is above 0.5.

At step 250, a sensor measurement record is ignored if a sensor reliability parameter associated with the sensor measurement record is below the sensor reliability threshold. As a result, assuming that the sensor reliability threshold is 0.5, all existence probability parameters along with their corresponding sensor measurement records corresponding to rows 3, 5, 7, and 8 of the existence probability table of 4 are ignored because values of the sensor reliability parameters in the first left cells (in column 1) of rows 3, 5, 7 and 8 are below 0.5.

At step 252, existence probability parameters of active paths that have not been completed and initialized paths are provided to the estimator, e.g., Kalman filter 106. At step 254, the estimator, e.g., Kalman filter 106, would then update a respective path existence value for each active path that has not been completed and each initialized path.

At step 256, the method 200 for associating object tracking data ends. It is to be understood, however, that the method 200 for associating object tracking data, which is performed in a plurality of sequential steps in 6 merely illustrates the method 200 for associating object tracking data as applied to a detection at a specific time, for example, in a specific frame of a video. As a result, step 256 may return to step 202 to re-perform the method 200 for associating object tracking data upon another detection at a different time after the specific time, for example, in a next frame of the video.

Thus, according to the object tracking data association method 200 using the object tracking data association module 100, it is now possible to effectively perform object tracking data association.

The LSTM network module 134 may be trained prior to its use in the method 200 for associating object tracking data. To generate labeled data sets for training the LSTM network module 134, values of the difference parameters of a training set may first be obtained, as described above. Then, difference parameters having values falling within a first range, e.g., 0 - 0.01, may be considered for use in association with paths that may be initialized.

Additionally, difference parameters having values falling within a second range, e.g., 0.01 - 0.2, may be considered for use in association with paths for which association may be performed. Further, difference parameters having values falling within a third range, e.g., 0.2 - 1, may be considered for use in association with paths for which association may be performed. As a result, a training set of labeled records may be generated for training the LSTM network module 134.

Although the invention has been described in some detail with respect to particular embodiments or aspects, other embodiments or aspects are possible.

Therefore, the spirit and scope of the appended claims should not be limited to the description of the embodiments contained herein.

For example, the at least one sensor reliability parameter may be obtained (step 238) and compared (step 240) in earlier steps in the method 200 for associating object tracking data.

All features disclosed in this specification (including the appended claims, abstract and accompanying drawings) may be replaced by alternative features serving the same or an equivalent or similar purpose, unless expressly stated otherwise. Accordingly, unless expressly stated otherwise, each feature disclosed is merely an example of a generic series of equivalent or similar features.

Claims (25)

Object tracking data association module (100), comprising: a path data retrieval device (110) configured to retrieve at least one path data set at least one path associated with at least one detected object; a sensor measurement data retrieval device (112) configured to provide at least one sensor measurement data set for the at least one path associated with the at least one detected object; and a difference parameter determination device (120) configured to determine at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object; wherein the difference parameter determination device (120) comprises an exponential function module (122) configured to perform calculations using an exponential function, and the difference parameter determination device (120) is configured to use the exponential function module (122) to calculate the at least one difference parameter using the exponential function. Object tracking data association module (100) according to Claim 1 , wherein the difference parameter determination device (120) has a correlation parameter determination device (124) which is designed to determine at least one correlation parameter which defines a correlation between the at least one sensor measurement data set and the at least one path data set. Object tracking data association module (100) according to Claim 2 , wherein the difference parameter determination device (120) has a statistical distance calculator (126) which is designed to statistically calculate the at least one correlation parameter. Object tracking data association module (100) according to one of the Claims 2 until 3 , wherein the exponential function has an exponent that has the at least one correlation parameter. Object tracking data association module (100) according to Claim 4 , wherein the exponent is a negative multiple of the at least one correlation parameter. Object tracking data association module (100) according to one of the Claims 2 until 3 , wherein the difference parameter determination device (120) comprises a Mahalanobis distance calculator (128) which is designed to calculate the at least one correlation parameter using a Mahalanobis distance formula, wherein the at least one correlation parameter comprises a Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set. Object tracking data association module (100) according to Claim 6 , wherein the exponential function has an exponent that represents the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set. Object tracking data association module (100) according to Claim 7 , wherein the exponent is a negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set. Object tracking data association module (100) according to one of the preceding claims, wherein the difference parameter determination device (120) is configured to calculate the at least one difference parameter by subtracting a value of the exponential function from one. Object tracking data association module (100) according to one of the preceding claims, further comprising an existence probability determination device (130) configured to accept the at least one difference parameter as input and to output at least one existence probability parameter for the at least one path associated with the at least one detected object. Object tracking data association module (100) according to Claim 10 , wherein the existence probability determination device (130) comprises a machine learning system (132) configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object. Object tracking data association module (100) according to one of the Claims 10 until 11 , wherein the existence probability determination device (130) comprises an LSTM network module (134) configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object. Object tracking data association module (100) according to one of the Claims 10 until 12 , wherein the path data retrieval device (110) is configured to provide at least one path data set comprising an active path data set of an active path associated with a previously detected object. Object tracking data association module (100) according to one of the Claims 10 until 13 , wherein the path data retrieval device (110) is configured to provide the at least one path data set comprising an inactive path data set of an inactive path associated with a newly recognized object. Object tracking data association module (100) according to one of the Claims 13 until 14 , further comprising a threshold comparator (140) configured to compare the at least one existence probability parameter of the at least one path with an existence threshold. Object tracking data association module (100) according to Claim 15 , further comprising a decision device (142) which is designed to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold. Object tracking data association module (100) according to Claim 15 , further comprising a decision device (142) which is designed to terminate the active path if the at least one existence probability parameter of the active path is below the existence threshold. Object tracking data association module (100) according to Claim 15 , further comprising a decision device (142) which is designed to initialize the inactive path to another active path if the at least one existence probability parameter of the inactive path is above the existence threshold. Object tracking data association module (100) according to one of the Claims 1 until 14 , further comprising a sensor reliability value retrieval device (114) configured to obtain at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set for the at least one path associated with the at least one detected object. Object tracking data association module (100) according to Claim 19 , further comprising a threshold comparator (140) configured to compare the at least one sensor reliability parameter with a sensor reliability threshold. Object tracking data association module (100) according to Claim 20 , further comprising a decision device (142) configured to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold. Object tracking module (102) comprising the object tracking data association module (100) according to any one of the preceding claims. Motor vehicle (108) comprising the object tracking data association module (100) according to one of the preceding claims. Object tracking data association module (100) according to one of the preceding claims, comprising: a path data retrieval device (110) configured to provide at least one path data set of at least one path associated with at least one detected object; wherein the path data retrieval device (110) is configured to provide the at least one path data set comprising an active path data set of an active path associated with a previously detected object; wherein the path data retrieval device (110) is configured to provide the at least one path data set comprising an inactive path data set of an inactive path associated with a newly detected object; a sensor measurement data retrieval device (112) configured to provide at least one sensor measurement data set for the at least one path associated with the at least one detected object; a difference parameter determination device (120) which is designed to determine at least one difference parameter which defines a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object; wherein the difference parameter determination device (120) has an exponential function module (122) which is designed to perform calculations using an exponential function, and the difference parameter determination device (120) is designed to use the exponential function module (122) to calculate the at least one difference parameter using application of the exponential function; wherein the difference parameter determination device (120) has a correlation parameter determination device (124) which is designed to determine at least one correlation parameter which defines a correlation between the at least one sensor measurement data set and the at least one path data set; wherein the difference parameter determination device (120) has a statistical distance calculator (126) which is designed to statistically calculate the at least one correlation parameter; wherein the difference parameter determination device (120) has a Mahalanobis distance calculator (128) which is designed to calculate the at least one correlation parameter using a Mahalanobis distance formula, wherein the at least one correlation parameter has a Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; wherein the difference parameter determination device (120) is designed to calculate the at least one difference parameter by subtracting a value of the exponential function from one; wherein the exponential function has an exponent that comprises the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; wherein the exponent has a negative multiple of the Mahalanobis distance between the at least one sensor measurement data set and the at least one path data set; an existence probability determination device (130) configured to accept the at least one difference parameter as input and output at least one existence probability parameter for the at least one path associated with the at least one detected object; wherein the existence probability determination device (130) comprises a machine learning system (132) configured to accept the at least one difference parameter as input and output the at least one existence probability parameter for the at least one path associated with the at least one detected object; wherein the existence probability determination device (130) comprises an LSTM network module (134) configured to accept the at least one difference parameter as input and to output the at least one existence probability parameter for the at least one path associated with the at least one detected object; a threshold comparator (140) configured to compare the at least one existence probability parameter of the at least one path with an existence threshold; a decision device (142) configured to perform an association for the active path if the at least one existence probability parameter of the active path is above the existence threshold; wherein the decision device (142) is configured to terminate an active path if the at least one existence probability parameter of the active path is below the existence threshold; wherein the decision device (142) is configured to initialize the inactive path to an active path if the at least one existence probability parameter of the inactive path is above the existence threshold; and a sensor reliability value retrieval device (114) configured to obtain at least one sensor reliability parameter associated with at least one sensor used to provide the at least one sensor measurement data set for the at least one path associated with the at least one detected object; wherein the threshold comparator (140) is configured to compare the at least one sensor reliability parameter to a sensor reliability threshold; wherein the decision device (142) is configured to ignore the at least one sensor measurement data set if the at least one sensor reliability parameter associated with the at least one sensor measurement data set is below the sensor reliability threshold. A method for associating object tracking data, comprising the following acts: providing, by a processor, at least one path data set of at least one path associated with at least one detected object; providing, by a processor, at least one sensor measurement data set for the at least one path associated with the at least one detected object; and determining, by a processor, at least one difference parameter defining a relationship between the at least one sensor measurement data set and the at least one path data set of the at least one path associated with the at least one detected object using an exponential function.

Images