Method, system, vehicle and medium for assigning vehicles to lanes

Abstract

The present invention relates to a method for assigning autonomous vehicles to lanes. In particular, it relates to a method and arrangement for assigning vehicles to lanes in a road for vehicles. The proposed solution obtains an estimated pose for the vehicle, projects the pose to nearby lanes, obtains uncertainty values ​​for the pose and lane, creates an instantaneous likelihood of the vehicle being in each lane by combining the pose and lane uncertainties, creates a distribution, adjusts the instantaneous likelihood with previous values ​​obtained in previous iterations, and determines the most likely lane.

Description

Technical Field

[0001] The present invention relates to a system, method and arrangement for determining the position of a vehicle, and in particular to a solution for determining in which lane a vehicle is located and assigning the vehicle to a lane. Background Art

[0002] In recent years, the technology for autonomous vehicles has developed by leaps and bounds, and many different solutions have been developed within sensor technology and control solutions for controlling the vehicle and determining objects in the surrounding area and the position of the vehicle relative to the road, other vehicles, objects, etc. to ensure safe travel of the vehicle. Today, development is ongoing in autonomous driving (AD) and advanced driver assistance systems (ADAS). Within this field there are several different technical areas. One such area is how to accurately and consistently locate the vehicle, as this is a basic safety feature for the vehicle when moving within the traffic flow. An accurate model of the road and its lanes is of fundamental importance for autonomous driving as the vehicle moves.

[0003] For determining the position of a mobile vehicle, relying solely on satellite-based positioning systems (Global Navigation Satellite System, GNSS), such as the Global Positioning System (GPS), the Global Navigation System (GLONASS), Galileo, BeiDou and other regional systems is often not accurate enough.

[0004] These GNSS based solutions are less precise in determining altitude / level information.

[0005] The ability to estimate the current autonomous lane based on the estimated position plays an important role when using a map to navigate and estimate the neighboring lane geometry. For example, if two roads are placed on top of each other, one turns left and the other turns right, the ability to estimate the lane geometry is directly related to the autonomous lane assignment. In addition, it has been found that using altitude information from GNSS is unstable - especially when driving underneath another road in situations where the received satellite signal is reduced - and does not provide the accuracy required to separate roads from each other.

[0006] Therefore, current solutions utilize a number of different sensors to increase the reliability of map positioning, such as image processing, LiDAR, vehicle travel parameters such as speed, etc. However, even when the position and instantaneous direction of the vehicle are known, it may still be difficult to predict or estimate the forward movement of the vehicle, such as in the above example where roads overlap, or when different road segment locations are close to each other. One such example is, for example, when the vehicle approaches a highway exit or a similar branch of the road.

[0007] Chinese patent document CN107221195 discusses a collision avoidance solution for vehicle lane prediction and lane-level mapping, in which a unique link ID number for each lane is used to determine the lane level, and the vehicle driving history is incorporated into the algorithm used to determine the current lane of the vehicle. Summary of the invention

[0008] It is an object to avoid at least some of the above disadvantages and to provide an improved node and method for assigning vehicles to lanes in a road.

[0009] This is provided in multiple embodiments, such as a method for assigning a vehicle to a lane in a road for the vehicle, wherein the method includes: obtaining an estimated pose E of the vehicle, obtaining uncertainty values ​​for the estimated pose of the vehicle, projecting the pose E onto nearby lanes in a map, obtaining position uncertainty U for each lane, creating a distribution P by combining the uncertainty V and the uncertainty U about each nearby lane, creating a current instantaneous likelihood hypothesis that the vehicle is in each lane by evaluating the distribution P with the pose E for each projection, adjusting the current instantaneous likelihood about each lane using a previously determined prior; comparing all adjusted instantaneous probabilities, and determining the lane with the highest likelihood.

[0010] In the method, the step of adjusting the current instantaneous possibility may include, for each possibility from a previous iteration: checking the logical link associated with the current instantaneous possibility, and if the connection is within a set distance from the pose E and if the current instantaneous possibility has not been adjusted previously, adjusting the current instantaneous possibility with the value of the possibility from the previous iteration; adjusting the current instantaneous possibility that was not adjusted in the previous step with the minimum value.

[0011] The set distance may be set to be shorter than 10 meters, or more preferably shorter than 5 meters, or even more preferably 2 meters or shorter.

[0012] The nearby lanes may be located within 20 meters of the estimated pose, or more preferably within 10 meters of the estimated pose, or even more preferably within 5 meters or less of the estimated position.

[0013] The method may further include the step of normalizing the current likelihoods. The method may further include the step of setting all likelihoods below a threshold to a fixed minimum likelihood value. The step of normalizing the current likelihoods may include determining the maximum value of the most likely current likelihoods and subtracting the maximum value from all current possibilities.

[0014] The maximum likelihood value for each lane is found at the center of the corresponding lane, and each lane probability value drops to 0 at a certain distance away from the center of the corresponding lane.

[0015] The step of adjusting the instantaneous likelihood may include examining logical connections between lane hypotheses in a map. The pose may include a location and a heading in map coordinates. Obtaining an estimated pose of the vehicle may include obtaining sensor data from at least one of a global navigation satellite system, a camera, vehicle parameters, and a lidar.

[0016] The method is advantageously implemented in an autonomous vehicle or an advanced driver assistance system vehicle.

[0017] Another embodiment is provided in a vehicle control system, the vehicle control system comprising at least one processor, at least one memory for storing instructions and data, at least one sensor interface, and at least one communication interface. The processor is arranged to: execute a set of instructions stored in the memory to obtain data via the sensor interface, and perform a method, namely, obtain an estimated pose E of the vehicle, obtain an uncertainty value V for the estimated pose of the vehicle, project the pose E onto nearby lanes in a map, obtain a position uncertainty U for each lane, create a distribution P by combining the uncertainty V and the uncertainty U for each nearby lane, create a current instantaneous probability hypothesis of the vehicle being in each lane by evaluating the distribution P with the pose E for each projection, adjust the current instantaneous probability for each lane with a previously determined prior; compare all adjusted instantaneous probabilities, determine the lane with the highest probability, and send control parameters for controlling the vehicle according to the determined most likely lane via the communication interface.

[0018] In the vehicle control system, the step of adjusting the current instantaneous likelihood may include, for each likelihood from a previous iteration: checking a logical connection associated with the current instantaneous likelihood, and if the connection is within a set distance from the pose E, and if the current instantaneous likelihood has not been previously adjusted, adjusting the current instantaneous likelihood by the value of the likelihood from the previous iteration; and adjusting the current instantaneous likelihood that was not adjusted in the previous step by the minimum value;

[0019] Another embodiment is provided, in which a vehicle includes a vehicle control system for controlling driving functions of the vehicle, the vehicle control system including: at least one processor, at least one memory for storing instructions and data, at least one sensor input interface, and at least one sensor for obtaining an estimate of the vehicle position, wherein the vehicle control system is arranged to execute instructions in a processing unit to perform the above method.

[0020] Yet another embodiment is provided, a computer program product arranged to perform the above method.

[0021] The proposed solution makes it possible to achieve an increased accuracy in determining which lane a vehicle is in. Using the proposed solution, more accurate position data can also be provided for local situations where many lanes occur and the lanes intersect each other.

[0022] This has the advantage of providing an efficient method of providing a high probability that the vehicle is in lane, thereby providing the system with the possibility of knowing the destination, determining a suitable onward route. This can be done in both the short and long term perspectives, i.e. at the operational and strategic level. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In the following, the invention will be described in more detail in a non-limiting manner with reference to exemplary embodiments illustrated in the accompanying drawings, in which:

[0024] Figure 1 is a schematic diagram illustrating an example vehicle;

[0025] Figure 2 is a schematic block diagram illustrating an exemplary control circuit;

[0026] Figure 3 is a schematic diagram illustrating an exemplary lane condition;

[0027] Figure 4 is a schematic diagram illustrating an exemplary probability distribution;

[0028] Figure 5 is a schematic block diagram illustrating an exemplary method; and

[0029] Figure 6 is a schematic block diagram illustrating an exemplary method. DETAILED DESCRIPTION

[0030] exist Figure 1 In the drawings, reference numeral 100 generally indicates a vehicle, or sometimes an autonomous vehicle, which may be autonomous, manually driven, or a combination of these, i.e., semi-autonomous. The vehicle 100 includes any suitable components for propelling and controlling the vehicle, such as an engine or electric motor, control electronics for controlling and monitoring the engine / motor, thrusters, steering wheels, gearboxes, brakes, lights, user interface dashboards, wheels, etc., as will be understood by those skilled in the art. However, for autonomous or semi-autonomous driving, the vehicle has Figure 2The electrical control circuit 210 is described in more detail, and one or more sensors 101, 102 for providing data such as data related to position, speed, direction of travel, temperature, road surface characteristics, etc. The position sensor may include a sensor that receives signals from one or more satellite-based positioning systems (for example, including from GPS satellites 104 via antenna 102), and / or from one or more cameras that provide images of the road and surrounding area. These images can be processed to determine the position of the vehicle relative to road boundaries, other vehicles, etc. The vehicle may also have an IMU (inertial measurement unit) (not shown) including accelerometers and gyroscopes for assisting in tracking the position and movement of the vehicle. Other sensors that may be used are, for example, LiDAR (light detection and ranging) and for tracking distance and speed relative to other vehicles around the vehicle; these sensors are not shown in the figure. Figure 1 1, but may be located within the vehicle, or incorporated into the sensor unit 101. The antenna may be used for communication with an external network and may also be used to receive GPS signals; however, separate antennas may be provided for external communications and GPS signals.

[0031] The vehicle may further be connected to an external network 150 via, for example, a wireless link 103. The same or some other wireless links may be used to communicate with other vehicles near the vehicle, or with local infrastructure elements. Cellular communication technology may be used for long-range communications such as to an external network, and if the cellular communication technology used has low latency, it may also be used for inter-vehicle, V2V, and / or vehicle-to-infrastructure, V2X communications. Examples of cellular radio technologies are GSM, GPRS, EDGE, LTE, 5G, 5G NR, etc., also including future cellular solutions. However, in some cases, medium and short-range communication technologies such as wireless local area networks (LANs), for example solutions based on IEEE 802.11, are used. ETSI is studying cellular standards for vehicle communications, for example, 5G is considered a suitable solution due to low latency and high bandwidth and efficient processing of communication channels.

[0032] The vehicle control circuit / system 210 is used to control the vehicle and Figure 2. The control circuit may include at least one processor 201 having a plurality of modules 250, 260 and 270 for performing different operations, and at least one computer readable memory or storage medium 202 for storing data and / or instruction sets for operating different functions in the processor. In addition, optionally, depending on the functions provided in the control circuit, one or more communication interfaces 205 and / or one or more antenna interfaces 207 may be provided, and in addition, one or more sensor interfaces 206 for acquiring data from sensors in the vehicle may also be provided. The modules of the processor may be a creation module 250, an adjustment module 260 and a comparison module 270, each arranged as an instruction set for executing a process of different steps for determining the autonomous lane of the vehicle. It should be understood that the communication / antenna interface may also provide the possibility of acquiring sensor data directly or via a proprietary sensor control circuit in the vehicle: for example, GPS data may be acquired through the antenna interface, and some sensors in the vehicle may communicate with the control circuit using a local network arrangement such as CAN bus, I2C, Ethernet, optical fiber, etc. The communication interface may be arranged to communicate with other control functions of the vehicle and thus may also be considered a control interface; however, a separate control interface (not shown) may be provided. Local communications within the vehicle may also be of a wireless type with protocols such as WiFi, LoRa, Zigbee, Bluetooth or similar medium / short range technologies.

[0033] The processor 201 may be, for example, a microprocessor, a digital signal processor, a graphics processing unit (GPU), an embedded processor, a field programmable gate array (FPGA), or an ASIC (application specific integrated circuit). The storage unit 202 may be a volatile or non-volatile computer readable memory, and is arranged to store instructions or instruction sets executed by the processor. The instruction set is preferably stored in a non-volatile memory, such as a solid state (SSD), a disk drive storage, an optical storage such as a CD, a DVD or a blue light, or a persistent solid state memory technology such as a flash memory or a memory card. The storage unit may also include a combination of storage types. The method may be implemented in a computer program product, and / or stored in a computer readable storage medium. The memory may be a non-temporary or temporary computer readable storage medium, which stores one or more programs configured to be executed by one or more processors of an electronic device with or without a display device and one or more input devices. The computer program product may be transferred on a storage medium such as, for example, an SSD, magnetic storage, optical storage, or may be transferred, for example, as a signal over a network connection (e.g., Ethernet using the Internet Protocol (IP)) using a suitable protocol, or wirelessly over a network connection using a suitable radio protocol such as cellular technology or short or medium range local area network technology.

[0034] Figure 3An exemplary lane situation where two roads 301, 302 are close to each other or partially or completely overlap is illustrated. In this case, each road includes two lanes, 310, 311, 312, 313, and the vehicle 100 travels on one lane 310. At a certain point when the vehicle travels along lane 310, it will be close to other lanes, such as lane 312, and the control circuit 100 needs to determine where the vehicle is traveling and which lane it is in at a specific time. Because of the uncertainty in determining the position, speed and yaw of the vehicle and the map accuracy, only the necessary parameters of limited accuracy are used to determine the position / pose of the vehicle in such a way that the vehicle in autonomous or semi-autonomous driving mode is safely controlled. The determination of the position of the vehicle is caused by uncertainty, such as noise in GPS signals, accuracy of speed measurement, accuracy of yaw measurement, camera resolution, delay problems during processing, map accuracy, etc. In one embodiment, parameters related to the pose of the vehicle are used, that is, the pose includes a two-dimensional position on the map and a driving direction. The two dimensions on the map are surface coordinates that exclude altitude information. However, it should be understood that the position data may also include altitude information which, when determining the position of the vehicle relative to the available lanes / roads, can be used in the following analysis to determine where the vehicle is located relative to the available lanes / roads.

[0035] Radar and lidar signals can also be used to determine vehicle location. For example, using radar and / or lidar maps of static objects such as signs, obstacles, buildings, etc. Known landmarks can also be used to locate the vehicle on the map; images of the surrounding area can be analyzed to identify objects and landmarks compared to known objects and landmarks.

[0036] exist Figure 4 In FIG. 4 , an example uncertainty U, 405 , 406 of the position of each lane 310 , 311 is depicted, which in this case may be, for example, a Gaussian distribution, such as Figure 4 As seen in , the likelihood decreases from the center of the lane 402, 401 on the road 301 toward the sides of the lane; however, other statistical distributions may be used to describe the uncertainty of the lane position. In a similar manner, different uncertainties of parameters may be determined or estimated. Figure 4 4, 5 and 6. It can be noted that the statistical distributions overlap 410, so there is a certain probability that a vehicle is in another lane than in the center of the distribution. The lane uncertainty U may depend on map uncertainty, lane size, etc. For example, in some areas, the map is more reliable than in other areas because it has been more thoroughly investigated, e.g., with multiple vehicles mapping these areas; this may be the case, for example, for high-speed traffic areas near cities or other densely populated areas. Some maps are built based on feedback from vehicles traveling in the area.

[0037] The processor in the processing system is arranged to iteratively execute the instruction set stored in the memory to obtain sensor data via the sensor interface and operate the method to assign the vehicle to the lane (so-called autonomous lane) by determining the lane where the vehicle is most likely to be located. This can be done by first creating 501 the instantaneous allocation probability of each nearby lane within a predetermined distance of the estimated position of the vehicle, adjusting 502 the instantaneous probability of each lane using a priori, comparing 503 all adjusted instantaneous probabilities, and determining the most likely lane, the prior being determined according to the previous iteration. The predetermined distance can be set to, for example, 5 meters, 10 meters or 20 meters, but it should be understood that other distances can be used, including shorter or longer distances. In the creation of the probability value, a road or road section that is logically connected to the position of the vehicle in the previous iteration will be used. Optionally, the processor can be arranged to send 504 control parameters for controlling the vehicle according to the geometry of the most likely lane determined by means of a communication / control interface. This process is iterated frequently enough to ensure safe determination of which lane the vehicle is in, for example at a frequency of 1-200 times per second, depending on vehicle speed and computing power.

[0038] The step of creating the instantaneous likelihood for the current iteration may include obtaining 601 an estimated pose E of the vehicle, i.e., the position on the map, and altitude information (which may or may not be available), and a direction of travel. The estimated pose may be obtained using sensor data such as data from a global navigation satellite system and / or based on determining fixed landmarks using other sensors. The estimated pose may be expressed in a global coordinate system or in a coordinate system based on a local map.

[0039] In addition, the method includes determining 602 an uncertainty value V for the estimated pose of the vehicle. The pose uncertainty can be expressed, for example, as a Gaussian around the most likely pose. The uncertainty V depends, for example, on sensor signals, map uncertainty, the update frequency of the sensors, the update frequency of the algorithm iterations, and the vehicle speed. V can be a set static value or a dynamic value determined at each iteration or at certain intervals.

[0040] E is projected 603 onto nearby lanes in the map, for example by finding the closest point from E to the nearby lanes. The projection for each lane advantageously also includes the lane direction. A distribution P is created 604 by merging the uncertainty V with the lane uncertainty U for each lane. The uncertainty of the lane can be represented as a Gaussian, which can depend on the map source and other factors that are static or dynamic values. The step of creating the distribution P can be performed by merging two Gaussians about V and the lane. Depending on which coordinate system is in and / or whether it is running in a decimal / logarithmic system, this merging of lane and pose uncertainties can be performed in different ways, for example by simply adding the two Gaussians to each other. The method further includes evaluating 605 the distribution P with the pose E, creating an instantaneous probability hypothesis that the vehicle is in each lane, and creating an instantaneous probability hypothesis that the vehicle is in each lane. The evaluation can be performed by subtracting E from each projection P and storing the results in a list.

[0041] Typically, this instantaneous likelihood for each lane is adjusted by a prior value determined from a previous iteration, for example if the vehicle was determined to be assigned to the lane in the last iteration then the prior value has a high value, but if the lane was not previously occupied the prior would be low.

[0042] For example, the system steps through all possibilities from the last iteration, and for each possibility (prior) from the previous iteration, the system determines in this step all logical connections to the current prior, i.e. the current possibility in the list of possibilities from the last iteration. If each logical connection associated to an instantaneous possibility is within a set distance to the vehicle pose and no prior was previously assigned to the logical connection, the connection is adjusted with the current prior possibility. This set distance may be, for example, 2 meters, 5 meters or 10 meters. It should be understood that the set distance is not limited to the illustrated distances, it may be larger or smaller. All instantaneous possibilities that were not adjusted above are adjusted with the minimum probability value. If there is a connection between lanes on the map, the connection is considered logical.

[0043] All lanes that have a logical connection or are within a certain distance of the nearest occupied lane will receive a non-zero value to indicate that there is some likelihood for the vehicle to be in that lane; however, for many lanes, the likelihood is very small. By comparing all adjusted instantaneous probabilities, it is possible to determine the lane that the vehicle is most likely to be in, i.e., determine the lane with the highest adjusted likelihood or probability.

[0044] By comparing the adjusted probabilities, the most likely lane can be found and the vehicle can be assigned to that lane.

[0045] Before using the likelihood values ​​in the next iteration, all non-zero values ​​are preferably normalized, for example so that all non-zero values ​​add up to a reference value, such as 1, 100%, or any other suitable value used in the statistical analysis to determine the total probability. Another way to normalize the likelihood values ​​is to subtract the maximum likelihood among the created possibilities for each possibility, i.e., if operating in a logarithmic scale, the normalized likelihood is equal to the unnormalized likelihood minus the maximum value of the unnormalized possibilities. This normalized probability value for each lane is set as the prior for that lane in the next iteration. For possibilities below a threshold, the likelihood value is set to the minimum likelihood value. For possible lanes with no prior value, the prior for that lane is set to a small non-zero value, such as -12 using a logarithmic scale when calculated. It should be understood that the normalization step can be performed before the comparison step.

[0046] The method in one implementation can be described in a short step-by-step example:

[0047] 1) Find the closest point (projection) from "E" to all nearby lanes, forming a set containing all projections

[0048] "projection_set". Each projection also contains lane direction information (lane heading).

[0049] 2) Create a distribution "P" (eg, by using a Gaussian distribution) based on the uncertainty "V" about "E" and the lane uncertainty.

[0050] 3) Create the instantaneous likelihood of each projection by evaluating (in "P" for each projection in projection_set (projection - "E").

[0051] 4) For each possibility (prior) from the previous iteration, starting from the highest and working downwards, do the following steps:

[0052] - Find all logical connections according to the current prior "CP"

[0053] - For each logical connection (including priors) associated to an instantaneous likelihood: if the connection is physically close to "E", and if no priors were added, adjust the instantaneous likelihood by "CP".

[0054] 5) Use the minimum likelihood value to adjust all instantaneous likelihoods that were not adjusted in step 4).

[0055] 6) Normalize all possibilities (possibility - maximum value of (possibility))) and set all possibility values ​​below the threshold to the minimum possibility value.

[0056] 7) Sort the list by likelihood, minus the most likely lane.

[0057] 8) Go to 1)

[0058] However, it should be understood that the order of some steps may be interchanged, for example, step 7 and step 6 may be performed in a different time sequence.

[0059] This approach is consistent with the Bainesian probability theory which operates with probability distributions.

[0060] The sensor and / or determined data may be low and / or high pass filtered in any suitable manner to reduce fluctuations, spurious data, and provide a smoother data set for controlling the vehicle.

[0061] The analysis can also be used to determine the appropriate direction of travel for the vehicle by predicting and determining the appropriate path forward knowing the current lane and destination. This can be used to determine the appropriate route on both a short-term and long-term basis, i.e. operational and strategic respectively. One such parameter used for planning is fuel / energy consumption.

[0062] In short: Given a map with attributes: logical connectivity between lanes, geometric description of centerlines, position of the autonomous vehicle relative to the map, a probabilistic autonomous lane assignment can be computed. First, a projection to all lanes in the vicinity of the autonomous vehicle is computed. Based on the uncertainty of the autonomous vehicle pose and the road, the distribution is evaluated and probabilities are assigned to all lanes for which a projection can be made. Second, the previous history of the lanes is exploited and the probabilities of lanes with priors are updated. All new lanes reachable from any previous lane have their probabilities updated with the most likely prior from which the lane is reachable, given the logical connectivity of the map. These two steps are then repeated iteratively, continuously updating the position of the vehicle relative to the road / lane.

[0063] The distribution used to evaluate the projection is at least a three-dimensional distribution, comprising a surface position (x, y) on a map and a heading (azimuth / yaw), ie the direction of travel.

[0064] With this solution, no lane height information is required to determine the most likely lane for a vehicle to occupy, reducing uncertainty in position and / or pose and enhancing safety.

[0065] It should be understood that parts of the described solutions may be implemented in the vehicle, in a system located external to the vehicle, or in a combination of internal and external to the vehicle; for example, in a server in communication with the vehicle, a so-called cloud solution. For example, sensor data may be sent to an external system, which performs the steps of determining the most likely lane, sending back information indicating the lane and other relevant parameters for controlling the vehicle.

[0066] Exemplary methods, apparatus, and computer-readable storage media are described in the following:

[0067] A method for assigning a vehicle (100) to a lane in a road (301), the method comprising:

[0068] - obtaining (601) an estimated position and pose of the vehicle;

[0069] - obtaining (602) an uncertainty value V for the estimated pose of the vehicle;

[0070] - Projecting (603) the pose E to a nearby lane in the map;

[0071] - Get the position uncertainty U for each lane;

[0072] - creating a distribution P by combining (604) the uncertainty V and the uncertainty U about each nearby lane;

[0073] - creating a hypothesis of the current instantaneous likelihood that the vehicle is in each lane by evaluating the distribution P with the pose E for each projection;

[0074] - adjusting (502) the current instantaneous likelihood for each lane using the previously determined priors;

[0075] - Comparing (503) all adjusted instantaneous probabilities and determining the lane with the highest probability.

[0076] According to one embodiment, the step of adjusting the current instantaneous likelihood may include, for each likelihood from a previous iteration:

[0077] - checking that a logical connection associated to said current instantaneous likelihood is within a set distance from pose E and adjusting the current instantaneous likelihood with the value of the likelihood from a previous iteration if the current instantaneous likelihood has not been adjusted previously; and

[0078] - adjust the current instantaneous probability that was not adjusted in the previous step by the minimum value;

[0079] According to one embodiment, the set distance may be set to be shorter than 10 meters, or more preferably shorter than 5 meters, or even more preferably 2 meters or shorter.

[0080] According to one embodiment, the nearby lanes may be located within 20 meters from the estimated pose, or more preferably within 10 meters from the estimated pose, or even more preferably within 5 meters or less from the estimated pose.

[0081] According to one embodiment, the method may further comprise the step of normalizing the current likelihood.

[0082] According to one embodiment, the method may further include setting all probabilities below a threshold to a fixed minimum probability value.

[0083] According to one embodiment, the step of normalizing the current possibilities may include determining a maximum value of the most likely current possibilities and subtracting this maximum value from all current possibilities.

[0084] According to one embodiment, the maximum value of the probability for each lane may be found at the center of the existing lane (401, 402), and each lane probability value may drop to zero at a certain distance away from the center of the corresponding lane.

[0085] According to one embodiment, the step of adjusting the instantaneous likelihood may include checking logical connections between lane hypotheses in the map.

[0086] According to one embodiment, the pose may include the position in map coordinates and the driving direction.

[0087] According to one embodiment, obtaining an estimated pose of a vehicle may include obtaining sensor (101, 102) data from at least one of a global navigation satellite system, a camera, vehicle parameters, and a lidar.

[0088] According to one embodiment, the vehicle may be an autonomous vehicle or an advanced driver assistance system vehicle.

[0089] A vehicle control system (210) comprises at least one processor (201), at least one memory (202) for storing instructions and data, at least one sensor interface (206), and at least one communication interface (205), wherein the processor is arranged to: execute an instruction set stored in the memory, obtain data via the sensor interface, and:

[0090] - obtaining (601) an estimated position E of the vehicle;

[0091] - obtaining (602) an uncertainty value V for the estimated pose of the vehicle;

[0092] - Projecting (603) the pose E onto a nearby lane in the map;

[0093] - Get the position uncertainty U for each lane;

[0094] - creating a distribution P by combining (604) the uncertainty V and the uncertainty U about each nearby lane;

[0095] - creating a hypothesis of the current instantaneous likelihood that the vehicle is in each lane by evaluating (605) the distribution P for each projection with the pose E;

[0096] - adjusting (502) the current instantaneous likelihood for each lane using the previously determined priors;

[0097] - sending (504) via the communication interface control parameters for controlling the vehicle according to the determined most probable lane.

[0098] According to one embodiment, the step of adjusting the current instantaneous likelihood may include, for each likelihood from a previous iteration:

[0099] - checking a logical connection associated to the current instantaneous likelihood and, if the connection is within a set distance from pose E and if the current instantaneous likelihood has not been previously adjusted, adjusting the current instantaneous likelihood with the value of the likelihood from a previous iteration; and

[0100] - Adjust the current instantaneous probability by the minimum value that was not adjusted in the previous step.

[0101] A vehicle (100) comprising

[0102] - a vehicle control system (210) for controlling driving functions of a vehicle, comprising at least one processor (201), at least one memory (202) for storing instructions and data, and at least one input interface (206);

[0103] - at least one sensor (101, 102) for obtaining an estimate of the vehicle's position;

[0104] Wherein the vehicle control system is arranged to execute instructions in the processing unit.

[0105] It should be noted that the word "comprising" does not exclude the presence of other elements or steps in addition to those listed, and the word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. It should be further noted that any reference signs do not limit the scope of the claims, the present invention may be implemented at least in part by hardware and software, and several "components" or "units" may be represented by the same hardware.

[0106] The embodiments mentioned and described above are given as examples only and should not be limiting to the present invention. For those skilled in the art, other solutions, uses, objectives and functions within the scope of the present invention as claimed in the patent embodiments described below should be obvious.

[0107] Abbreviations

[0108] WiFi Wireless Fidelity (a registered trademark of the WiFi Alliance) - based on IEEE 802.11 protocol

[0109] LoRa Long Range Radio (registered trademark of the LoRa Alliance)

[0110] Zigbee Wireless communication protocol for IoT and similar applications, controlled by the Zigbee Alliance

[0111] GSM Global System for Mobile Communications

[0112] GPRS General Packet Radio Service

[0113] EDGE Global Evolution Enhanced Data Rate

[0114] LTE Long Term Evolution

[0115] IP Internet Protocol

[0116] Radar Radio Detection and Ranging

[0117] Lidar Light Detection and Ranging

Claims

1. A method for assigning a vehicle (100) to a lane in a road (301), wherein the method comprises: Obtaining (601) an estimated position E of the vehicle; Obtaining (602) an uncertainty value V for the estimated pose of the vehicle; Projecting (603) the pose E onto a nearby lane in the map; Get the position uncertainty U of each lane; Creating a distribution P by combining (604) the uncertainty value V and the position uncertainty U for each nearby lane; Creating a current instantaneous likelihood hypothesis of the vehicle being in each lane by evaluating (605) the distribution P for each projection with the pose E; adjusting (502) the current instantaneous likelihood for each lane using the previously determined priors; and All adjusted instantaneous probabilities are compared (503) and the lane with the highest probability is determined.

2. The method of claim 1, wherein the step of adjusting the current instantaneous likelihood comprises: For each possibility from the previous iteration: checking a logical connection associated with the current instantaneous likelihood and, if the connection is within a set distance from the pose E and if the current instantaneous likelihood has not been previously adjusted, adjusting the current instantaneous likelihood with the value of the likelihood from the previous iteration; and The current instantaneous likelihood that was not adjusted in the previous step is adjusted by the minimum likelihood value. The method according to claim 2 , wherein the set distance is set to be shorter than 10 meters. The method of claim 1 , wherein nearby lanes are located within 20 meters of the estimated pose.

5. The method of claim 1, further comprising the step of normalizing the current instantaneous likelihood.

6. The method of claim 5, further comprising the step of setting all probabilities below a threshold to a fixed minimum probability value.

7. The method of claim 5, wherein the step of normalizing the current likelihood comprises: The maximum value of the most likely current possibilities is determined and this maximum value is subtracted from all current possibilities.

8. The method according to any one of claims 1 to 7, wherein the maximum value of the probability for each lane is found at the corresponding lane center (401, 402), and each lane probability value drops to 0 at a certain distance away from the center of the corresponding lane.

9. The method of claim 2, wherein the step of adjusting the instantaneous probability comprises: Check the logical connections between the lane assumptions in the map.

10. The method of claim 1, wherein the pose comprises a location in map coordinates and a driving direction.

11. The method of claim 1, wherein obtaining the estimated pose of the vehicle comprises obtaining sensor (101, 102) data from at least one of a global navigation satellite system, a camera, vehicle parameters, and a lidar.

12. The method of claim 1, wherein the vehicle is an autonomous vehicle or an advanced driver assistance system vehicle.

13. A vehicle control system (210), comprising at least one processor (201), at least one memory (202) for storing instructions and data, at least one sensor interface (206), and at least one communication interface (205), wherein the processor is arranged to: execute a set of instructions stored in the memory, acquire data via the sensor interface, and: Obtaining (601) an estimated position E of the vehicle; Obtaining (602) an uncertainty value V for the estimated pose of the vehicle; Projecting (603) the pose E to a nearby lane in the map; Get the position uncertainty U of each lane; Creating a distribution P by combining (604) the uncertainty value V and the position uncertainty U for each nearby lane; Creating a current instantaneous likelihood hypothesis of the vehicle being in each lane by evaluating (605) the distribution P for each projection with the pose E; adjusting (502) the current instantaneous likelihood for each lane using the previously determined priors; comparing (503) all adjusted instantaneous probabilities and determining the lane with the highest probability; as well as Control parameters for controlling the vehicle according to the determined most likely lane are sent (504) via the communication interface.

14. The vehicle control system of claim 13, wherein the step of adjusting the current instantaneous likelihood comprises, for each likelihood from a previous iteration: checking a logical connection associated with the current instantaneous likelihood and, if the connection is within a set distance from the pose E and if the current instantaneous likelihood has not been previously adjusted, adjusting the current instantaneous likelihood with the value of the likelihood from the previous iteration; and The current instantaneous likelihood that was not adjusted in the previous step is adjusted by the minimum likelihood value.

15. A vehicle (100), comprising: A vehicle control system (210) for controlling the driving function of the vehicle, comprising at least one processor (201), at least one memory (202) for storing instructions and data, and at least one sensor input interface (206); at least one sensor (101, 102) for obtaining an estimate of a vehicle's position; Wherein the vehicle control system is arranged to execute instructions in the processing unit to perform a method according to any one of claims 1 to 7 or claims 9 to 12.

16. A computer program product arranged to perform the method according to any one of claims 1 to 7 or 9 to 12.

Images