GB2634053A

GB2634053A - Road type detection apparatus and method for detecting a type of road currently taken by an ego vehicle

Info

Publication number: GB2634053A
Application number: GB2314880.2A
Authority: GB
Inventors: Venkatesh Kambakana; Naga Venkateswara Rao Kaja; Sravani Dokku
Original assignee: Continental Autonomous Mobility Germany GmbH
Current assignee: Aumovio Autonomous Mobility Germany GmbH
Priority date: 2023-09-28
Filing date: 2023-09-28
Publication date: 2025-04-02
Also published as: GB202314880D0; WO2025068052A1

Abstract

Road type estimation is achieved by gathering environmental parameters and dynamic parameters of an ego vehicle. For each of at least two different types of roads probability values are calculated indicating a probability that a respective type of road is currently being traversed. The probabilities are obtained by a calculation using the detected parameters. A temporary result is determined during each sampling interval indicating a type of road the vehicle is assumed to take based on a comparison of each of the probability values for each of the at least two different types of roads. A final result is determined by incrementing, during each sampling interval, a respective counter value for one respective type of road with which the type of road indicated by the temporary result matches, and setting the respective type of road as the final result, the counter value of which exceeded a specified threshold value relevant for the respective type of road.

Description

Road type detection apparatus and method for detecting a type of road currently taken by an ego vehicle

TECHNICAL FIELD

The invention relates to a road type detection apparatus and computer-implemented method for detecting a type of road currently taken by a vehicle. Moreover, the invention relates to a driver assistance system for a vehicle, comprising such a road type detection apparatus, a corresponding computer program 10 for carrying out the computer-implemented method and a respective non-transitory computer-readable medium.

BACKGROUND

Conventional Driver Assistance Systems or Advanced Driver Assistance Systems (ADAS) for vehicles are widely used, for example, in connection with conventional driver-assisted vehicles as well as autonomous vehicles, and usually comprise applications such as an Adaptive Cruise Control (ACC), Traffic-Sign-Assist (TSA), Speed-Limit-Assist (SLA) and so on, which applications aim to avoid vehicle accidents that may have been caused by human errors.

For example, the Traffic Sign Assist application ensures that the current speed limit and other road signs are displayed to the driver on an ongoing basis, for example, on a screen installed inside the vehicle.

Automatic recognition functions via a link between images captured by a camera or image sensors, for example, a front camera mounted on a front of the vehicle, and the speed limit information stored in the navigation system. In this way, even speed limits that are not explicitly visible, like within a city, may be displayed to the driver.

Optionally, the Speed-Limit-Assist application can be extended to an Intelligent Speed Assist (ISA) by adapting the cruise control with speed limit information from traffic sign recognition.

ADAS applications are also used in connection with road type estimation. This is because knowledge of the type of road the ego vehicle currently takes is quite often relevant for an adequate cruise control in terms of vehicle speed control and so on.

However, in the prior art, only a limited number of input parameters to decide on road type like signs, lanes, and so on is used for the road type estimation (RTE).

For example, the non-patent literature "Road-Type Detection Based on Traffic Sign and Lane Data" by Zoltan Fazekas, Gabor Balazs Csaba Gyulai, Peter Potyondi, and Peter Gaspar, deals, Hindawi, Journal of Advanced Transportation, Volume 2022, Article ID 6766455, deals with RTE on the basis of such a limited number of input parameters.

Further prior art is constituted by the documents US 2002128751 Al and EP 3875906 Al.

However, there is a problem that using a limited number of input parameters for road type detection leads to incorrect estimations in a few obvious situations, and that due to inaccuracies in detection, a kind of flickering occurs on the screen when different road types are repeatedly displayed in complicated scenarios.

Since an accurate road type estimation is crucial to display a correct speed type for an ego vehicle and for a suitable cruise control, the prior art algorithms are quite often not suitable for use in dynamic situations.

SUMMARY OF THE INVENTION

A road type detection apparatus according to the invention is for detecting a type of road currently taken by an ego vehicle (i.e. the ego vehicle being equipped with said road type detection apparatus), wherein the road type detection apparatus comprises: parameter acquisition means configured to acquire, during each sampling interval, a plurality of currently detected parameters including at least one or more environmental parameters currently detected in respect of the environment of the ego vehicle and one or more dynamic parameters currently detected in respect of the dynamics of the ego vehicle, road type probability value calculation means configured to calculate, for each of at least two different types of roads and during each sampling interval, probability values indicating a probability that a respective type of road is currently taken by the ego vehicle, each respective probability value for a respective type of road being obtained by a calculation using currently detected parameters relevant for the respective type of road, acquired by the parameter acquisition means, temporary result determining means configured to determine, during each sampling interval, a temporary result indicating one type of road the ego vehicle is assumed to take based on a comparison of each of the probability values for each of the at least two different types of roads, and final result determining means configured to determine a final result indicating the result of detection of one type of road the ego vehicle currently takes by (i) incrementing, during each sampling interval, a respective counter value for one respective type of road with which the type of road indicated by the temporary result matches, and (ii) setting the respective type of road as the final result, the counter value of which exceeded a specified threshold value relevant for the respective type of road.

One advantageous effect achieved by the road type detection apparatus according to the invention is that RTE accuracy (road type estimation accuracy) is enhanced due to considering a plurality of currently detected parameters including at least one or more environmental parameters currently detected in respect of the environment of the ego vehicle and one or more dynamic parameters currently detected in respect of the dynamics of the ego vehicle in order to decide on the type of road. Moreover, the temporary result determining means and final result determining means form some kind of confidence building algorithm to avoid flickering of estimated value, i.e. to increase the probability of estimating the correct type of road. The invention can be employed in any camera based ADAS application and serves finding the relevant road type in order to provide accurate applicable speed limits for the ego vehicle.

Accordingly, as multiple inputs are used to estimate the road type in connection with the confidence building algorithm, estimated output is stabilized. The road type detection apparatus according to the invention can be further modified in such a manner that the final result determining means is further configured to (iii) reset each counter value for each respective type of road if one type of road is set as the final result, and/or (iv) retain the setting of the one respective type of road as the final result until a counter value for another type of road exceeds a specified threshold value relevant for the other type of road.

Accordingly, as long as a current final result in respect of a specific type of road is set, this final result will remain until a counter value of another type of road reaches the specified threshold value. This prevents any flickering as accuracy of the road typed estimation is increased.

Moreover, the road type detection apparatus according to the invention can be further realized such that the final result determining means is further configured to (v) determine the final result indicating no detection of one type of road the ego vehicle currently takes until a respective counter value for one respective type of road exceeds a specified threshold value for the first time.

Furthermore, the road type detection apparatus according to the invention can be further configured in such a manner that the one or more currently detected parameters include at least one of the environmental parameters selected from the group of - vehicle lane information being information on a detected number of lanes adjacent to the lane the ego vehicle currently takes, -speed limit indicated by a speed limit sign detected to be located in the road currently taken by the ego vehicle, - vehicle lane change information being information on a detected number of lane changes during a predetermined past time period or past distance traveled by the ego vehicle, -distance between last turn taken by ego vehicle and the ego vehicle, - number of vehicles, preferably the number of trucks, passed per past distance traveled by the ego vehicle, - pedestrian information being information on a detected number of pedestrians detected in a predetermined past time period or past distance traveled by the ego vehicle, - number of city signs detected in a predetermined past time period or past distance traveled by the ego vehicle, - oncoming traffic information being information on a number of vehicles that have approached the ego vehicle, -play street signs information being information on whether a play street sign has been detected in a predetermined past time period or past distance traveled by the ego vehicle (10), and at least one dynamic parameter being -the vehicle speed being the detected current speed of the ego vehicle.

Accordingly, those parameters which are collected by means of image sensors or camera sensors are collected for every frame captured by the image sensors or camera sensors.

For example, parameters like the vehicle lane information, the speed limit (information), the vehicle lane change information, the distance between last turn taken by ego vehicle and the ego vehicle, the number of vehicles, preferably the number of trucks, the pedestrian information, the number of city signs, the oncoming traffic information, and the play street signs information may be detected by image or camera sensors and obtained by a respective processing known from the prior art.

The at least one dynamic parameter being the vehicle speed may be detected by a conventional vehicle speed sensor known in the prior art.

Beyond that, the road type detection apparatus according to the invention can be further implemented in such a manner that the at least two different types of roads are selected from the group of (i) motorway, (ii) city road, (iii) country road, and (iv) play street.

The term "motorway" is preferably understood as high speed allowing road or roads connecting different countries, and meant for travelling long distances.

For example, a motorway (also referred to as freeway or expressway) may be a controlled-access highway which is a type of highway that has been designed for high-speed vehicular traffic, with all traffic flow, i.e. ingress and egress, regulated.

The term "city road" is preferably understood as road or roads internal to a city, i.e. that run through a city.

The term "country road" is preferably understood as rural area road or roads. The term "playstreet" is preferably understood as a community road or roads so as to provide a space for school age children to participate in recreational activities.

As will be apparent below, the road type detection apparatus according to the invention determines the type of road based on an evaluation of specific parameters being used criteria for the classification of the several types of roads.

In this connection, the road type detection apparatus can be further modified such that the road type probability value calculation means is configured to (i) calculate a probability value for the type of road being the motorway using the parameters including - the vehicle lane information, - the speed limit, -the vehicle lane change information, - the distance between last turn taken by ego vehicle and the ego vehicle, - the number of vehicles, preferably the number of trucks, passed per past distance traveled by the ego vehicle, and - the vehicle speed, and/or (ii) calculate a probability value for the type of road being the city road using the parameters including - the speed limit, - the distance between last turn taken by ego vehicle and the ego vehicle, - the pedestrian information, -the number of city signs, and - the vehicle speed, and/or (iii) calculate a probability value for the type of road being the country road using the parameters including - the vehicle lane information, -the vehicle lane change information, - the oncoming traffic information, and - the vehicle speed, and/or (iv) calculate a probability value for the type of road being the play street using the parameters including -the vehicle speed, - the pedestrian information, and - the play street signs information.

The aforementioned criteria, i.e. parameters, selected for calculating each specific a probability value for each type of road have been determined on the basis of empirical data.

Moreover, the road type detection apparatus according to the invention can be further realized in that the road type probability value calculation means is configured to (i) calculate the probability value for the motorway by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weight, wherein -the vehicle speed, the vehicle lane information, and the speed limit are weighted with the first weight, - the distance between last turn taken by ego vehicle and the ego vehicle is weighted with the second weight, and - the vehicle lane change information and the number of vehicles, preferably 15 the number of trucks, passed by the ego vehicle per distance traveled by the ego vehicle are weighted with the third weight, and/or (ii) calculate the probability value for the city road by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weigh, wherein -vehicle speed and pedestrian information are weighted with the first weight, - speed limit and number of city signs are weighted with the second weight, and - distance between last turn taken by ego vehicle and the ego vehicle is weighted with the third weight, and, and/or (iii) calculate the probability value for the country road by weighting each of the relevant parameters with the same weight, and/or (iv) calculate the probability value for the play street by weighting each of the relevant parameters with the same weight.

The aforementioned weights for the respective criteria, i.e. parameters, have 30 been determined based on empirical data.

Furthermore, the road type detection apparatus according to the invention can be further configured in such a manner that one sample interval corresponds to the period of time in which one frame is detected by detecting means being at least one image sensor or camera for capturing an image of the environment of the ego vehicle, the captured image corresponding to a frame. That is, the detection apparatus works on currently captured frames obtained/detected by the image sensor or camera.

Additionally, the road type detection apparatus according to the invention can be further configured such that the temporary result determining means is configured to determine the temporary result to be the type of road for which the probability value is calculated to be maximum with respect to the probability values for the other respective types of roads. Thus, only the type of road is selected as temporary result, which has the maximum probability value with respect to the other remaining probability values of the other types of roads.

Moreover, the road type detection apparatus according to the invention can be further implemented in such a manner that the road type probability value calculation means is configured to calculate probability values which take values between 0 and 1.

A driver assistance system for an ego vehicle according to the invention com prises: detecting means configured to detect, during each sampling interval, a plurality of parameters including one or more environmental parameters detected in respect of the environment of the ego vehicle and one or more dynamic parameters detected in respect of the dynamics of the ego vehicle, a road type detection apparatus according to the invention as described above, using the parameters detected by the detecting means, and displaying means configured to display the final result, obtained from the road type detection apparatus.

Accordingly, the properties and advantages explained in connection with the road type detection apparatus according to the invention arise in the same or similar manner in respect of the driver assistance system according to the invention, which is why, in order to avoid repetitions, reference is made to the respective explanations with respect to the road type detection apparatus according to the invention.

The driver assistance system according to the invention can be further modified in such a manner that the detecting means comprises at least one image sensor or camera for capturing an image of the environment of the ego vehicle, each image being captured during one sample interval.

A computer-implemented method according to the invention is for detecting a type of road currently taken by an ego vehicle, wherein the method comprises the following steps: a parameter acquisition step of acquiring, during each sampling interval, a plurality of currently detected parameters including at least one or more environmental parameters currently detected in respect of the environment of the ego vehicle and one or more dynamic parameters currently detected in respect of the dynamics of the ego vehicle, road type probability value calculation step of calculating, for each of at least two different types of roads and during each sampling interval, probability values indicating a probability that a respective type of road is currently taken by the ego vehicle, each respective probability value for a respective type of road being obtained by a calculation using currently detected parameters relevant for the respective type of road, acquired in the parameter acquisition step, temporary result determining step of determining, during each sampling interval, a temporary result indicating one type of road the ego vehicle is assumed to take based on a comparison of each of the probability values for each of the at least two different types of roads, and final result determining step of determining a final result indicating the result of 20 detection of one type of road the ego vehicle currently takes by (i) incrementing, during each sampling interval, a respective counter value for one respective type of road with which the type of road indicated by the temporary result matches, and (ii) setting the respective type of road as the final result, the counter value of which exceeded a specified threshold value relevant for the respective type of road.

Accordingly, the properties and advantages explained in connection with the road type detection apparatus according to the invention arise in the same or similar manner in respect of the computer-implemented method according to the invention, which is why, in order to avoid repetitions, reference is made to the respective explanations with respect to the road type detection apparatus according to the invention.

A computer program according to the invention comprises instructions which, when the program is executed by a computer, cause the computer to carry out the computer-implemented method according to the invention.

A non-transitory computer-readable medium according to the invention has 5 stored thereon the computer program according to the invention.

BRIEF DESCRIPTION OF THE DRAWINGS

Preferred embodiments and advantages of the invention will become more apparent upon reading the following detailed description along with the accompanying drawings in which: FIG. 1 shows a flowchart of an overall procedure for detecting a type of road currently taken by an ego vehicle according to one embodiment of the invention; FIG. 2 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a motorway; FIG. 3 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a city road; FIG. 4 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a country road; FIG. 5 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a play street; FIG. 6 shows a flowchart of a procedure for determining the type of road currently taken by the vehicle; and FIG. 7 shows a configuration of road type detection apparatus according to one embodiment of the invention, included in a driver assistance system of a vehicle.

DETAILED DESCRIPTION

In this context, the road type detection apparatus as described in this description may include a memory which is for example used in the processing carried out in the road type detection apparatus. A memory used in the embodiments may be a volatile memory, for example a DRAM (Dynamic Random Access Memory) or a non-volatile memory, for example a PROM (Programmable Read Only Memory), an EPROM (Erasable PROM), EEPROM (Electrically Erasable PROM), or a flash memory, e.g., a floating gate memory, a charge trapping memory, an MRAM (Magnetoresistive Random Access Memory) or a PCRAM (Phase Change Random Access Memory).

In this context, the road type detection apparatus as described in this description may include a processor or a "circuit". A "circuit" may be understood as any kind of a logic implementing entity, which may be special purpose circuitry or a processor executing software stored in a memory, firmware, or any combination thereof. Thus, in an embodiment, a "circuit" may be a hard-wired logic circuit or a programmable logic circuit such as a programmable processor, e.g. a microprocessor (e.g. a Complex Instruction Set Computer (CISC) processor or a Reduced Instruction Set Computer (RISC) processor). A "circuit" may also be a processor executing software, e.g. any kind of computer program, e.g. a computer program using a virtual machine code such as e.g. Java. Any other kind of implementation of the respective functions which will be described in more detail below may also be understood as a "circuit" in accordance with an alternative embodiment.

With reference to the drawings, one preferred embodiment of the present invention will now be specifically described for illustrative purposes. It is to be understood that components or elements in this embodiment will be shown and described by way of examples only, but are not intended to limit this invention to them.

A road type detection apparatus 100 according to the invention is for detecting a type of road currently taken by an vehicle 10 (also referred to as "ego vehicle" meaning that the ego vehicle is equipped with said road type detection apparatus 100), in which the road type detection apparatus 100 is provided.

First, the structure/configuration of the road type detection apparatus 100 is described before the operation of the same is discussed in more detail.

FIG. 7 shows a configuration of the road type detection apparatus 100 according to one embodiment of the invention, which is included in a driver assistance system 1 of the (ego) vehicle 10.

As can be gathered from FIG. 7, the driver assistance system 1 is included in the (ego) vehicle 10, which may be -in this embodiment -a driver assisted vehicle or an autonomous vehicle, wherein the driver assistance system 1, in turn, comprises detecting means 60, the road type detection apparatus 100 according to the invention and displaying means 70.

The detecting means 60 is configured to detect, during each sampling interval, a plurality of parameters including one or more environmental parameters detected in respect of the environment of the vehicle 10 and one or more dynamic parameters detected in respect of the dynamics of the ego vehicle 10. For example, the detecting means 60 comprises one or more image sensors or camera sensors for capturing an image of the environment of the ego vehicle, each image being captured during said one sample interval.

In this embodiment, one sample interval corresponds to the period of time in which one frame is detected by detecting means being the image sensor or camera for capturing an image of the environment of the ego vehicle, the captured image corresponding to a frame.

The road type detection apparatus 100 according to invention will be described below in more detail in connection with its operation, which is a computer-implemented method according to the invention. A computer program comprising instructions which, when the program is executed by a computer, may cause the computer to carry out the computer-implemented method according to the invention, wherein a non-transitory computer-readable medium may store thereon the computer program.

The displaying means 70 is configured to display the final result obtained by the road type detection apparatus 100. For example, displaying means 70 is a screen inside the vehicle to display information to a driver of the vehicle 10.

As can be further gathered from FIG. 7, the road type detection apparatus 100 comprises parameter acquisition means 20, road type probability value calculation means 30, temporary result determining means 40 and final result determining means 50, the functioning of each will be described in more concrete terms below in connection with the operation of the road type detection apparatus 100.

FIG. 1 shows a flowchart of an overall procedure for detecting a type of road 30 currently taken by an ego vehicle 10 according to one embodiment of the invention, carried out by the road type detection apparatus 100.

In step S11, for every frame, a list of parameters is collected. This is carried out by the parameter acquisition means 20 which is configured to acquire, during each sampling interval, a plurality of currently detected parameters including some environmental parameters currently detected in respect of the environment of the ego vehicle 10 and one dynamic parameter currently detected in respect of the dynamics of the ego vehicle 10.

In this respect, the currently detected environmental parameters are: -vehicle lane information being information on a detected number of lanes adjacent to the lane the ego vehicle 10 currently takes, - speed limit indicated by a speed limit sign detected to be located in the road currently taken by the ego vehicle 10, - vehicle lane change information being information on a detected number of lane changes during a predetermined past time period or past distance traveled by the ego vehicle 10, - distance between (a position of a) last turn taken by ego vehicle 10 and (the current position of) the ego vehicle 10, - number of trucks, passed per past distance traveled by the ego vehicle 10, -pedestrian information being information on a detected number of pedestrians detected in a predetermined past time period or past distance traveled by the ego vehicle 10, - number of city signs detected in a predetermined past time period or past distance traveled by the ego vehicle 10, -oncoming traffic information being information on a number of vehicles that have approached the ego vehicle 10 (i.e. the number of vehicles belonging to the opposing or oncoming traffic), - play street sign information being information on whether a play street sign has been detected in a predetermined past time period or past distance traveled by the ego vehicle 10.

Moreover, the one dynamic parameter taken into consideration in this embodiment is the vehicle speed being the detected current speed of the ego vehicle 10. However, further dynamic parameters may be taken into consideration, which are, for example, engine speed, angle at the steering wheel if required, and so on.

If all parameters are acquired in step S11, the procedure moves to step S12, in which an indicators analysis for road type estimation takes place.

In other words, the respective procedures as shown in FIGS. 2 to 5 for indicating the respective probabilities that the road currently taken by the vehicle is a specific road type are carried out.

In this embodiment, the road type detection apparatus 100 is configured to distinguish between four different types of roads, which are a motorway, city road, country road, and play street.

The road type probability value calculation means 30 is configured to calculate, for each of these different types of roads and during each sampling interval, probability values indicating a probability that a respective type of road is currently taken by the ego vehicle 10, wherein each respective probability value for a respective type of road is obtained by a calculation using only those currently detected parameters which are relevant for the respective type of road, acquired by the parameter acquisition means 20.

FIG. 2 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a motorway.

As can be gathered from FIG. 2, in step S21, the road type probability value calculation means 30 collects relevant parameters for calculating a probability value for the type of road being the motorway. The relevant parameters are in this case the vehicle lane information, the speed limit, the vehicle lane change information, the distance between last turn taken by ego vehicle 10 and the ego vehicle 10, the number of vehicles, in this case the number of vehicles in the form of trucks, passed per past distance traveled by the ego vehicle 10, and the vehicle speed, wherein the road type probability value calculation means 30 calculates the probability value for the motorway by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weight.

In the case of FIG. 2, the vehicle speed, the vehicle lane information, and the speed limit are weighted with the first weight, the distance between last turn taken by ego vehicle 10 and the ego vehicle 10 is weighted with the second weight, and the vehicle lane change information and the number of vehicles in the form of trucks passed by the ego vehicle 10 per distance traveled by the ego vehicle 10 are weighted with the third weight.

Thus, in this way, in step S22, the road type probability value calculation means 30 calculates and outputs a probability value for the motorway that takes a value between 0 and 1.

FIG. 3 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a city road.

As can be gathered from FIG. 3, in step S31, the road type probability value calculation means 30 collects relevant parameters for calculating a probability value for the type of road being the city road. The relevant parameters are in this case the speed limit, the distance between last turn taken by ego vehicle and the ego vehicle, the pedestrian information, the number of city signs, and the vehicle speed, wherein the road type probability value calculation means 30 calculates the probability value for the city road by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weigh. In this case, the vehicle speed and pedestrian information are weighted with the first weight, the speed limit and number of city signs are weighted with the second weight, and the distance between last turn taken by ego vehicle 10 and the ego vehicle 10 is weighted with the third weight.

Thus, in this way, in step S32, the road type probability value calculation means 30 calculates and outputs a probability value for the city road that takes a 15 value between 0 and 1.

FIG. 4 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a country road.

As can be gathered from FIG. 4, in step S41, the road type probability value calculation means 30 collects relevant parameters for calculating a probability value for the type of road being the country road. The relevant parameters are in this case the vehicle lane information, the vehicle lane change information, the oncoming traffic information, and the vehicle speed, wherein the road type probability value calculation means 30 calculates the probability value for the country road by weighting each of the relevant parameters with the same weight.

Thus, in this way, in step S42, the road type probability value calculation means 30 calculates and outputs a probability value for the country road that takes a value between 0 and 1.

FIG. 5 shows a flowchart of a procedure for indicating the probability that the road currently taken by the vehicle is a play street.

As can be gathered from FIG. 5, in step S51, the road type probability value calculation means 30 collects relevant parameters for calculating a probability value for the type of road being the play street. The relevant parameters are in this case the vehicle speed, the pedestrian information, and the play street signs information, wherein the road type probability value calculation means 30 calculates the probability value for the play street by weighting each of the relevant parameters with the same weight.

Thus, in this way, in step S52, the road type probability value calculation means 30 calculates and outputs a probability value for the play street that takes a 5 value between 0 and 1.

Turning back to FIG. 1, after having calculated and output the probability values for each type of road, the procedure moves to step S13, in which confidence building algorithm is carried out by a confidence building module including the temporary result determining means 40 and final result determining means 50.

FIG. 6 shows a flowchart of a procedure for determining the type of road currently taken by the vehicle, corresponding to the confidence building algorithm. In step S61, the temporary result determining means 40 determines, during each sampling interval, a temporary result indicating one type of road the ego vehicle 10 is assumed to take based on a comparison of each of the probability values for each of the different types of roads, wherein temporary result is determined to be the type of road for which the probability value is calculated to be maximum with respect to the probability values for the other respective types of roads.

Dependent on which probability value is maximum, the procedure moves from one of steps S62a, S62b, S62c and S62d to one of steps S63a, S63b, S63c and S63d, in which the final result determining means 50 increments, during each sampling interval, a respective confidence counter value C_MW, C_CY, C_PS or C_CR for the one respective type of road with which the type of road indicated by the temporary result matches, by a threshold value (threshold), for example by 1.

The procedure moves then to step S64, where it is determined whether one of 25 the confidence counter values C_MW, C_CY, C_PS or C_CR exceeds a respective confidence threshold (S641a, S641b, S641c and S641d).

If one of the confidence counter values C_MW, C_CY, C_PS or C_CR exceeds a respective confidence threshold, the procedure moves to the respective one of steps S642a, S642b, S642c and S642, and the respective type of road is set as the final result, while the remaining other counter values for the remaining other types of roads are reset.

Example:

The confidence counter values C_MW reaches a value greater than 100 for 100 frames, i.e. during 100 sampling intervals. In this case, when C_MW reaches the value 101 and the confidence threshold for Motorway is 100, the final result is set to be motorway as road type and the all confidence counters will be reset then.

If none of the confidence counter values C_MW, C_CY, C_PS or C_CR exceeds said respective confidence threshold, the procedure moves to steps S65.

In step S65, if one type of road is set as the final result, the setting of the one respective type of road as the final result is retained until a confidence counter value for another type of road exceeds a specified confidence threshold value relevant for the other type of road.

If no type of road is set as the final result, a final result is set, indicating no detection of one type of road the ego vehicle 10 currently takes until a respective counter value for one respective type of road exceeds a specified threshold value for the first time.

Accordingly, another road type will only be set if a respective counter value of another road type exceeds the confidence threshold. This process avoids flickering in displaying estimated road types.

The features of the invention disclosed in the foregoing description, in the drawings as well as in the claims may be essential for the realization of the invention both individually and in any combination.

List of Reference Signs 1 driver assistance system vehicle (ego vehicle) 20 parameter acquisition means road type probability value calculation means temporary result determining means final result determining means detecting means 70 displaying means road type detection apparatus

Claims

Claims 1. A road type detection apparatus (100) for detecting a type of road currently taken by an ego vehicle (10), preferably the apparatus being configured to carry out the method according to claim 13, the road type detection apparatus (100) comprising: parameter acquisition means (20) configured to acquire, during each sampling interval, a plurality of currently detected parameters including at least one or more environmental parameters currently detected in respect of the environment of the ego vehicle (10) and one or more dynamic parameters currently detected in respect of the dynamics of the ego vehicle (10), road type probability value calculation means (30) configured to calculate, for each of at least two different types of roads and during each sampling interval, probability values indicating a probability that a respective type of road is currently taken by the ego vehicle (10), each respective probability value for a respective type of road being obtained by a calculation using currently detected parameters relevant for the respective type of road, acquired by the parameter acquisition means (20), temporary result determining means (40) configured to determine, during each sampling interval, a temporary result indicating one type of road the ego vehicle (10) 20 is assumed to take based on a comparison of each of the probability values for each of the at least two different types of roads, and final result determining means (50) configured to determine a final result indicating the result of detection of one type of road the ego vehicle (10) currently takes by (i) incrementing, during each sampling interval, a respective counter value for one respective type of road with which the type of road indicated by the temporary result matches, and (ii) setting the respective type of road as the final result, the counter value of which exceeded a specified threshold value relevant for the respective type of road.
2. The road type detection apparatus (100) according to claim 1, wherein the final result determining means (50) is further configured to (iii) reset each counter value for each respective type of road if one type of road is set as the final result, and/or (iv) retain the setting of the one respective type of road as the final result until a counter value for another type of road exceeds a specified threshold value relevant for the other type of road.
3. The road type detection apparatus (100) according to claim 1 or 2, wherein final result determining means (50) is further configured to (v) determine the final result indicating no detection of one type of road the ego vehicle (10) currently takes until a respective counter value for one respective type of road exceeds a specified threshold value for the first time.
4. The road type detection apparatus (100) according to claim 1 to 3, wherein 15 the one or more currently detected parameters include at least one of the environmental parameters selected from the group of - vehicle lane information being information on a detected number of lanes adjacent to the lane the ego vehicle (10) currently takes, - speed limit indicated by a speed limit sign detected to be located in the road currently taken by the ego vehicle (10), - vehicle lane change information being information on a detected number of lane changes during a predetermined past time period or past distance traveled by the ego vehicle (10), - distance between last turn taken by ego vehicle (10) and the ego vehicle 25 (10), - number of vehicles, preferably the number of trucks, passed per past distance traveled by the ego vehicle (10), - pedestrian information being information on a detected number of pedestrians detected in a predetermined past time period or past distance traveled by the ego vehicle (10), - number of city signs detected in a predetermined past time period or past distance traveled by the ego vehicle (10), - oncoming traffic information being information on a number of vehicles that have approached the ego vehicle (10), - play street signs information being information on whether a play street sign has been detected in a predetermined past time period or past distance traveled by the ego vehicle (10), and at least one dynamic parameter being -the vehicle speed being the detected current speed of the ego vehicle (10).
5. The road type detection apparatus (100) according to claim 1 to 4, wherein the at least two different types of roads are selected from the group of (i) motorway, (ii) city road, (iii) country road, and (iv) play street.
6. The road type detection apparatus (100) according to claim 5, when dependent on claim 4, wherein the road type probability value calculation means (30) is configured to (i) calculate a probability value for the type of road being the motorway using the parameters including - the vehicle lane information, -the speed limit, - the vehicle lane change information, - the distance between last turn taken by ego vehicle (10) and the ego vehicle (10), - the number of vehicles, preferably the number of trucks, passed per past distance traveled by the ego vehicle (10), and - the vehicle speed, and/or (ii) calculate a probability value for the type of road being the city road using the parameters including - the speed limit, -the distance between last turn taken by ego vehicle and the ego vehicle, - the pedestrian information, - the number of city signs, and - the vehicle speed, and/or (iii) calculate a probability value for the type of road being the country road using the parameters including - the vehicle lane information, - the vehicle lane change information, -the oncoming traffic information, and - the vehicle speed, and/or (iv) calculate a probability value for the type of road being the play street using the parameters including - the vehicle speed, -the pedestrian information, and - the play street signs information.
7. The road type detection apparatus (100) according to claim 6, wherein the road type probability value calculation means (30) is configured to (i) calculate the probability value for the motorway by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weight, wherein - the vehicle speed, the vehicle lane information, and the speed limit are weighted with the first weight, -the distance between last turn taken by ego vehicle (10) and the ego vehicle (10) is weighted with the second weight, and - the vehicle lane change information and the number of vehicles, preferably the number of trucks, passed by the ego vehicle (10) per distance traveled by the ego vehicle (10) are weighted with the third weight, and/or (ii) calculate the probability value for the city road by weighting the relevant parameters with a first weight, a second weight smaller than the first weight, and a third weight smaller than the second weigh, wherein - vehicle speed and pedestrian information are weighted with the first weight, - speed limit and number of city signs are weighted with the second weight, 30 and - distance between last turn taken by ego vehicle (10) and the ego vehicle (10) is weighted with the third weight, and, and/or (iii) calculate the probability value for the country road by weighting each of the relevant parameters with the same weight, and/or (iv) calculate the probability value for the play street by weighting each of the relevant parameters with the same weight.
8. The road type detection apparatus (100) according to any one of claims 1 to 7, wherein one sample interval corresponds to the period of time in which one frame is detected by detecting means being at least one image sensor or camera for capturing an image of the environment of the ego vehicle, the captured image corresponding to a frame.
9. The road type detection apparatus (100) according to any one of claims 1 to 8, wherein the temporary result determining means (40) is configured to determine the temporary result to be the type of road for which the probability value is calculated to be maximum with respect to the probability values for the other respective types of roads.
10. The road type detection apparatus (100) according to any one of claims 1 to 9, wherein road type probability value calculation means (30) is configured to calculate probability values which take values between 0 and 1.
11. A driver assistance system (1) for an ego vehicle (10), comprising: detecting means (60) configured to detect, during each sampling interval, a plurality of parameters including one or more environmental parameters detected in respect of the environment of the ego vehicle (10) and one or more dynamic parameters detected in respect of the dynamics of the ego vehicle (10), a road type detection apparatus (100) according to any of claims 1 to 10, and displaying means (70) configured to display the final result.
12. The driver assistance system (1) according to claim 11, wherein the detecting means (60) comprises at least one image sensor or camera for capturing 30 an image of the environment of the ego vehicle, each image being captured during one sample interval.
13. A computer-implemented method for detecting a type of road currently taken by an ego vehicle (10), preferably carried out by a road type detection apparatus (100) according to any one of the claims 1 to 10, wherein the method comprises the following steps: a parameter acquisition step of acquiring, during each sampling interval, a plurality of currently detected parameters including at least one or more environmental parameters currently detected in respect of the environment of the ego vehicle (10) and one or more dynamic parameters currently detected in respect of the dynamics of the ego vehicle (10), road type probability value calculation step of calculating, for each of at least two different types of roads and during each sampling interval, probability values indicating a probability that a respective type of road is currently taken by the ego vehicle (10), each respective probability value for a respective type of road being obtained by a calculation using currently detected parameters relevant for the respective type of road, acquired in the parameter acquisition step, temporary result determining step of determining, during each sampling interval, a temporary result indicating one type of road the ego vehicle is assumed to take based on a comparison of each of the probability values for each of the at least two different types of roads, and final result determining step of determining a final result indicating the result of detection of one type of road the ego vehicle currently takes by (i) incrementing, during each sampling interval, a respective counter value for one respective type of road with which the type of road indicated by the temporary result matches, and (ii) setting the respective type of road as the final result, the counter value of which exceeded a specified threshold value relevant for the respective type of road.
14. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the computer-implemented method of claim 13.
15. A non-transitory computer-readable medium having stored thereon the computer program of claim 14.