System, control unit and method for determining geo-fence events for a vehicle

Abstract

The present invention provides a system, control unit and method for deciding a geo-fence event of a vehicle. The control unit includes: a first road information generation module, which is configured to generate first road information based on received map data and vehicle position data, which at least includes one or more candidate lanes based on the vehicle position; a second road information generation module, which is configured to generate second road information based on received radar data, which at least includes a detection lane based on the radar data; a calculation module, which is configured to perform a fusion calculation on the first road information and the second road information to obtain the confidence of each candidate lane, and determine the lane where the vehicle is located based on the calculated confidence; and a decision module, which is configured to decide whether to trigger a geo-fence event based on the determined lane, which includes an event of the vehicle entering the geo-fence and an event of leaving the geo-fence.

Description

Technical Field

[0001] The present invention relates to a system for deciding a geo-fence event of a vehicle and a control unit thereof, and also relates to a method for deciding a geo-fence event. Background Art

[0002] With the advancement of network communication technology, autonomous driving solutions have developed from single-vehicle intelligence to the collaboration between smart vehicles and smart roads, which has greatly promoted the development of smart cities.

[0003] Geofencing will play an important role in the development of smart cities because currently vehicles equipped with driver assistance systems and autonomous driving functions mainly reduce traffic accidents by enhancing the driver's senses. However, with the development of vehicle automation and vehicle networking, in order to improve the safety of the traffic environment, it is necessary to combine vehicle technology with geofencing technology, that is, to perform autonomous driving with the help of geofencing.

[0004] The existing solution of combining vehicle technology with geo-fencing technology mainly relies on map data, positioning data based on navigation satellites and camera data. However, the roads for autonomous driving with the help of electronic fences usually include restricted roads such as elevated roads, expressways, and tunnels. For these restricted roads, the use of positioning data and camera data based on navigation satellites has great instability factors, because the positioning signal based on navigation satellites will weaken or even disappear in these road areas, resulting in positioning failure. The camera will collect environmental information in situations such as strong light sources, ground water reflections, and rain-induced lens distortion. This will cause the vehicle to make wrong decisions about geo-fencing and autonomous driving, and may further cause serious traffic accidents.

[0005] Therefore, an improved technical solution is urgently needed to solve the above problems in the prior art. Summary of the invention

[0006] In view of the above problems in the prior art, the present invention aims to provide a vehicle control solution that is capable of deciding geo-fence events for the vehicle.

[0007] According to an embodiment of the first aspect of the present invention, a control unit for deciding a geo-fence event of a vehicle is provided, comprising: a first road information generation module, configured to generate first road information based on received map data and vehicle position data, which at least includes one or more candidate lanes based on the vehicle position; a second road information generation module, configured to generate second road information based on received radar data, which at least includes a detected lane based on the radar data; a calculation module, configured to perform a fusion calculation on the first road information and the second road information to obtain a confidence level of each candidate lane, and determine the lane in which the vehicle is located based on the calculated confidence level; and a decision module, configured to decide whether to trigger a geo-fence event based on the determined lane decision, the geo-fence event including an event of a vehicle entering a geo-fence and an event of a vehicle leaving a geo-fence.

[0008] According to an embodiment of the second aspect of the present invention, a system for deciding geo-fence events for a vehicle is provided, comprising: a networked device configured to receive map data for autonomous driving; a positioning device configured to receive a vehicle positioning signal and generate vehicle position data; a radar sensor configured to capture road information of a road ahead of the vehicle and generate radar data; and a control unit as described above, configured to perform a fusion calculation on the map data, the vehicle position data and the radar data, and decide whether to trigger geo-fence events for the vehicle, wherein the geo-fence events include events of the vehicle entering a geo-fence and events of the vehicle leaving a geo-fence.

[0009] According to an embodiment of the third aspect of the present invention, a method for deciding a geo-fence event for a vehicle is provided. Optionally, the method is executed by a control unit as described above and / or a system as described above, and the method includes: generating first road information based on received map data and vehicle position data, which at least includes one or more candidate lanes based on the vehicle position; generating second road information based on received radar data, which at least includes a detected lane based on the radar data; fusing the first road information and the second road information to obtain the confidence of each candidate lane, and determining the lane in which the vehicle is located based on the calculated confidence; and deciding whether to trigger a geo-fence event based on the determined lane, the geo-fence event including an event of a vehicle entering a geo-fence and an event of a vehicle leaving a geo-fence.

[0010] According to an embodiment of a fourth aspect of the present invention, there is provided a machine-readable medium on which machine-readable instructions are stored. When the instructions are executed, the machine is caused to execute the method as described above.

[0011] It can be seen that according to the embodiment of the present invention, through the fusion processing of map data and radar data, the misjudgment of the lane in which the vehicle is located is eliminated, the stability and accuracy of decision-making geo-fence events are improved, and the cost is low. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a schematic block diagram of a system for deciding a geo-fence event of a vehicle according to a feasible embodiment of the present invention, which includes a control unit according to a feasible embodiment of the present invention.

[0013] Figure 2 An exemplary process of deciding a geo-fence event according to a feasible embodiment of the present invention is schematically illustrated.

[0014] Figure 3 The calculation principle of the first confidence level according to a possible implementation of the present invention is schematically shown.

[0015] Figure 4 The calculation principle of the second and third confidence levels according to a possible embodiment of the present invention is schematically shown.

[0016] Figure 5 is a flow chart of a method for deciding a geo-fence event for a vehicle according to a possible embodiment of the present invention. DETAILED DESCRIPTION

[0017] Considering the various obstacles in the prior art regarding decision-making geofence events, the inventors found that the types of roads included in geofences are usually isolated using physical metal and cement as media, such as guardrails and soundproof walls on both sides of the road, which happen to be features that radars are good at detecting. The inventors also found that by using the on-board radar equipped for the vehicle's driving assistance system (especially the millimeter-wave radar, which is not expensive but performs extremely well in detecting metal guardrails, lamp poles, bridges, etc.), the radar data and the map data of the high-precision map used for autonomous driving can be detected to determine the lane in which the vehicle is located, thereby making a decision on whether to trigger a geofence event. The inventors also designed a fusion calculation method for radar data and map data. Using such a fusion calculation method, the lane in which the vehicle is located can be efficiently and stably identified, thereby improving the efficiency and accuracy of decision-making.

[0018] The embodiments of the present invention are mainly related to a vehicle control scheme for deciding geo-fence events for a vehicle.Before describing the embodiments of the present invention, the terms "geofence" and "geofence event" appearing in the specification and claims are explained.

[0019] In the embodiments of the present invention, "geofencing" can be understood as a virtual boundary based on a real map, and its attributes and shape can be dynamically generated and defined. The geofence can be implemented as an invisible boundary defined by coordinates such as global positioning satellite (GPS) coordinates. The present invention does not limit the method for generating the geofence.

[0020] In the embodiment of the present invention, a "geo-fence event" may be understood to include an event in which a vehicle enters a geo-fence area and an event in which a vehicle leaves a geo-fence area.

[0021] According to an embodiment of the present invention, after the event of a vehicle entering a geo-fence is triggered, the vehicle starts autonomous driving (for example, autonomous driving above level L2) and executes a series of control strategies for autonomous driving (for example, environmental perception, behavior planning, trajectory prediction, etc.).

[0022] According to an embodiment of the present invention, after the event of the vehicle leaving the geo-fence is triggered, the vehicle ends the autonomous driving and allows the driver to take over the vehicle.

[0023] Therefore, accurate decision-making on geo-fencing events plays a vital role in the safety of vehicle autonomous driving.

[0024] It is understood that "entering a geo-fence" may include two situations where the vehicle is about to enter the geo-fence and is about to enter the geo-fence. The timing of the vehicle entering the geo-fence may be defined by a predetermined time and / or a predetermined distance, that is, how long and / or how far the vehicle will travel before entering the geo-fence. In this way, the vehicle can be prepared to start autonomous driving.

[0025] It is understood that "leaving the geofence" may include two situations: the vehicle is about to leave the geofence and is about to leave the geofence. The timing of the vehicle leaving the geofence can be defined by a predetermined time and / or a predetermined distance, that is, how long and / or how far the vehicle will travel before leaving the geofence. In this way, the vehicle can notify the driver in advance and prepare to take over the vehicle.

[0026] Hereinafter, some embodiments of the present invention are described with reference to the accompanying drawings.

[0027] Figure 1 A system 100 for deciding a geo-fence event of a vehicle according to a feasible embodiment of the present invention is schematically shown, which mainly includes: a control unit 10 , a networking device 20 , a positioning device 30 and a radar sensor 40 .

[0028] The networking device 20 is arranged on the vehicle and can realize the networking functions of the vehicle, such as V2X communication (vehicle-to-everything communication), V2I communication (vehicle-to-infrastructure communication) and V2V communication (vehicle-to-vehicle communication). Figure 1 The networking device 20 can be connected to the server 1 via the network 2. The networking device 20 is used to receive a map MAP for autonomous driving from the server 1.

[0029] In one embodiment, server 1 may be implemented as a map cloud for autonomous driving, in which a map MAP for autonomous driving, such as a high-precision map, is stored. Networked device 20 may obtain a map MAP for autonomous driving, for example, by downloading the map MAP from a map cloud via network 2. In this embodiment, network 2 is typically implemented as a wireless network, which may be based on any radio communication technology and / or standard. For example, network 2 may include a telecommunications network of any standard provided by a telecommunications operator. Although Figure 1 A single network 2 is shown in FIG. 4 , but the network 2 may be configured to include a plurality of networks.

[0030] In another embodiment, the network 2 may also include the Internet of Things (IoT). For example, the network 2 may be implemented as a self-organizing wireless network. A vehicle with networking capabilities may, as a node, have the function of discovering other nodes with which it can establish a self-organizing wireless network. Other nodes may include roadside facilities, mobile terminals, and the like within the communicative range of the vehicle. Examples of such a network 2 may include, but are not limited to, a wireless local area network (WLAN), an infrared (IR) network, a Bluetooth network, a near field communication (NFC) network, a ZigBee network, and the like. In this embodiment, the networking device 20 may obtain a map MAP from a roadside server or an edge server storing a map MAP via the network 2.

[0031] The positioning device 30 can locate the vehicle by means of a positioning signal from a vehicle navigation system and generate vehicle position data including the vehicle position. The positioning device 30 can be implemented as a vehicle-mounted GPS, for example.

[0032] The radar sensor 40 is arranged on the vehicle to detect the road information in front of the vehicle and generate radar data. The radar data may include radar data of multiple radar detection points (radar wave reflection points). The radar sensor 40 may include one or more of a laser radar, a millimeter wave radar, and an ultrasonic radar. In one embodiment, the radar sensor 40 is implemented as a long-range millimeter wave radar arranged at the front of the vehicle.

[0033] The control unit 10 is used to execute such a vehicle control strategy, that is, to perform a fusion calculation based on the map data, the vehicle position data and the radar data and identify the lane in which the vehicle is located, and then decide whether to trigger a geo-fence event based on the determined lane. In the process of deciding whether to trigger a geo-fence event, (1) map auxiliary information obtained from the map data; and / or (2) radar auxiliary information obtained from the radar data can be considered to assist in deciding the geo-fence event.

[0034] The control unit 10 may be provided in an electronic control unit (ECU) of the vehicle, that is, the control unit 10 may be implemented by means of an electronic control unit (ECU) in the vehicle.

[0035] Continue to see Figure 1 , the control unit 10 can be connected in communication with the vehicle's autonomous driving control unit 3 (i.e., a control unit for a driver assistance system) to transmit the decision on the geo-fence event to the autonomous driving control unit 3. After receiving the decision, the autonomous driving control unit 3 can execute the autonomous driving control strategy. For example, in response to the decision of entering the geo-fence event, the autonomous driving control unit 3 performs behavior planning and route prediction.

[0036] The control unit 10 can be implemented in software or hardware or a combination of software and hardware. In one embodiment, the control unit 10 includes a memory and a processor. The memory contains instructions, which, when executed by the processor, at least cause the processor to perform the above-mentioned fusion calculation and make the above-mentioned decision.

[0037] In this embodiment, the memory has multiple implementations, for example, it can be implemented as a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical storage device, a magnetic storage device, or a suitable combination thereof.

[0038] In this embodiment, the processor has multiple implementations, for example, it can be implemented as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or a suitable combination thereof. The general-purpose processor can be a microprocessor, a controller, or a microcontroller. The processor can also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, a combination of one or more microprocessors and a DSP core, or other similar structures.

[0039] Continue to see Figure 1The control unit 10 mainly includes a first road information generating module 11, a second road information generating module 12, a calculation module 13 and a decision module 14. These modules can be implemented as software modules. One or more of these modules can be combined or further divided into multiple sub-modules according to specific application conditions.

[0040] It is understandable that the naming of the modules 11-14 of the control unit 10 should be understood as a logical description rather than a limitation of the physical form or setting method. In other words, one or more of the modules 11-14 can be implemented in the same chip or circuit, or they can be respectively set in different chips or circuits, and the present invention does not limit this. For the modules of the control unit 10, as long as an on-board control unit has their functions, it should be understood that the on-board control unit includes these modules.

[0041] The following, combined Figure 1-Figure 4 The working principle of the control unit 10 will be described below.

[0042] Figure 2 An exemplary process 200 for deciding a geo-fence event according to a possible implementation of the present invention is schematically shown. The process 200 may be executed in the control unit 10 .

[0043] The first road information generation module 11 receives (block 210 ) map data and vehicle position data, and generates (block 230 ) first road information based on the map data and the vehicle position data. The map data may come from a high-precision map for autonomous driving. The vehicle position data may come from a positioning device 30 .

[0044] The first road information at least includes one or more candidate lanes based on the vehicle position. In one embodiment, the module 11 proposes one or more candidate lanes within the predetermined search range (e.g., the area defined by a pre-set 10m*10m square search box) in the high-precision map MAP based on the vehicle position and the predetermined search range. The module 11 can also calculate the boundaries of each candidate lane based on the map data.

[0045] In addition, when no road or lane is found within the predetermined search range, the search range may be expanded (for example, a predetermined 10m*10m search box may be expanded to a 20m*20m search box) until a candidate road is found.

[0046] The first road information may also include map auxiliary information. For example, after determining the candidate lanes, the module 11 may extract information related to the candidate lanes from the high-precision map as the map auxiliary information.

[0047] The map auxiliary information may include one or more of the following: (1) road grade (e.g., expressway / expressway / urban expressway / national highway / provincial highway / county / township road); (2) road type (e.g., main road / secondary road); (3) lane type (e.g., normal lane / emergency lane); and (4) boundary type and shape of the candidate lane (e.g., boundary type may include material information of the boundary guardrail of the candidate lane, which may indicate whether a boundary is made of a medium such as cement or metal that can be detected by a radar sensor, and shape may include geometric shape information of the candidate lane, such as a straight lane, a curved lane, and the intersection relationship between multiple lanes). This information can assist in making decisions on vehicle geofencing events.

[0048] It is understandable that although the vehicle position can be known based on the vehicle position data, the vehicle position may not be very accurate, especially in scenarios such as parallel lanes, intersecting lanes, and elevated roads, where lane positioning may deviate. Therefore, the following control strategy is required to eliminate lane positioning deviations, thereby accurately and stably knowing the lane in which the vehicle is located.

[0049] The second road information generation module 12 receives the radar data (block 220) and generates (block 240) the second road information based on the radar data. The radar data may come from the radar sensor 40 and include distance information and / or speed information detected by the radar sensor 40 within its detection range. The radar data may include data of multiple radar detection points (reflection points).

[0050] The second road information at least includes the edge of the detected lane based on the radar data. In one embodiment, the module 12 fits multiple radar detection points into a linear pattern as the boundary of the detected lane (for example, guardrails, concrete piers, soundproof walls, etc. at the road boundary).

[0051] The second road information may also include radar auxiliary information. The radar auxiliary information may include speed information based on radar data. For example, the radar sensor 40 may capture the speed information of the vehicle on the lane and the surrounding lanes, and the module 12 calculates the average speed of each lane based on the speed information.

[0052] It is understandable that the present invention does not limit the order in which the control unit 10 receives the radar data, the map data, and the vehicle position data, and these data can be received by the control unit 10 at the same time or at respective predetermined cycles. In other words, the blocks 210 and 220 can be executed at the same time or at respective predetermined cycles.

[0053] Next, the calculation module 10 performs a fusion calculation on the first road information and the second road information to obtain the confidence of each candidate lane, and determines the lane in which the vehicle is located based on the calculated confidence (block 250 ).

[0054] In the embodiment of the present invention, the confidence of a candidate lane can be understood as the degree of matching between the candidate lane and the detected lane. In other words, the higher the confidence of the candidate lane, the greater the possibility that the candidate lane is determined to be the lane where the vehicle is located.

[0055] The calculation module 10 may consider two factors, namely, the confidence level and the confidence threshold, to determine the lane the vehicle is in. The confidence threshold may be determined based on one or more factors in experience, model calculation, or actual vehicle testing.

[0056] In the case where the candidate lanes include only one lane, when the calculated confidence of the lane is greater than or equal to the confidence threshold, the lane is determined as the lane where the vehicle is located. When the calculated confidence of the lane is less than the confidence threshold, it is considered that the lane where the vehicle is located has not been found. In the case where the lane where the vehicle is located has not been found, new data can be received and the confidence can be continuously calculated based on the new data as the vehicle changes dynamically (the vehicle moves), until the lane where the vehicle is located is found.

[0057] In the case where the candidate lanes include multiple lanes, the lane with the highest calculated confidence and the highest confidence greater than or equal to the confidence threshold is determined as the lane where the vehicle is located. In other words, even if the lane with the highest confidence is selected, but the highest confidence is still less than the confidence threshold, it is considered that the lane where the vehicle is located has not been found. In the case where the lane where the vehicle is located has not been found, new data can be received as the vehicle moves and the confidence can be continuously calculated based on the new data until the lane where the vehicle is located is found.

[0058] The confidence of the candidate lanes can be calculated based on multiple types of confidence. In one embodiment, the confidence of the candidate lanes can be calculated based on one or more of the following: (1) a first confidence associated with the shortest distance between the vehicle position and each candidate lane, for example, the first confidence can be calculated based on the shortest distance between the vehicle position and each candidate lane; (2) a second confidence associated with the distance between the vehicle position and each radar detection point, for example, the second confidence can be calculated based on the distance between the vehicle position and each radar detection point; and (3) a third confidence associated with the minimum distance between the detection point on the boundary of the detection lane and the boundary of each candidate lane, for example, the third confidence can be calculated based on the minimum distance between the detection point on the boundary of the detection lane and the boundary of each candidate lane.

[0059] In one embodiment, the confidence of the candidate lane can be calculated based on the following formula:

[0060] C i =W 1 *C i1+W 2 *C i2 +W 3 *C i3 ,

[0061] Among them, C i represents the confidence of the i-th candidate lane; C i1 represents the first confidence of the i-th candidate lane, W 1 Represents C i1 The weight of C i2 represents the second confidence of the i-th candidate lane, W 2 Represents C i2 The weight of C i3 represents the third confidence of the i-th candidate lane, W 3 Represents C i3 The weight of .

[0062] In this embodiment, the weight C of each confidence level is i1～ C i3 It is adjustable. For example, the weights of different types of confidence can be adjusted according to specific application scenarios.

[0063] In the following, some calculation examples of the first, second and third confidence levels are described in the form of examples.

[0064] In one embodiment of calculating the first confidence, the calculation module 13 calculates the shortest distance between the vehicle position and the center line of each candidate lane in the lateral direction of the vehicle, and determines the first confidence based on the shortest distance. The shorter the shortest distance between the vehicle position and the center line of a candidate lane, the higher the first confidence of the candidate lane. The center line of each candidate lane can be obtained from a high-precision map.

[0065] See also Figure 3 , the first confidence of the first candidate lane L1 is determined based on the shortest distance R1 between the position of the vehicle V (e.g., the center or centroid of the vehicle V) and the centerline L1' of the lane L1. Similarly, the first confidence of the second candidate lane L2 is determined based on the shortest distance R2 between the position of the vehicle V and the centerline L2' of the lane L2. The first confidence of the third candidate lane L3 is determined based on the shortest distance R3 between the position of the vehicle V and the centerline L3' of the lane L3. Figure 3 As shown, the candidate lanes L1-L3 may include curved lanes, and the candidate lanes L1-L3 may also intersect each other. Therefore, as the vehicle V travels, the first confidence of each candidate lane changes.

[0066] It is understandable that the present invention does not limit how to calculate the shortest distance between the vehicle and the center of the lane. For a straight lane or a curved lane, any suitable calculation method can be adopted to achieve it.

[0067] In one embodiment of calculating the second confidence level, the distance between each detection point on the radar detection line and the current position of the vehicle is calculated, and the second confidence level is determined based on the distance. The larger the distance, the lower the second confidence level.

[0068] In another embodiment of calculating the second confidence level, the second confidence level can be calculated by setting a confidence interval. For example, taking the current position of the vehicle as the starting point, multiple confidence intervals are set along the vehicle heading, and confidence levels are allocated to each confidence interval. For the allocation of confidence levels, one way is: the farther the confidence interval is from the vehicle in the vehicle heading, the lower the confidence level of the confidence interval. Another way is: in each confidence interval, the distance between each radar detection point and the vehicle position is counted, and the confidence level of the confidence interval is allocated according to the counted distance.

[0069] See also Figure 4 , three confidence intervals are divided along the vehicle heading, namely, the first confidence interval AREA① (for example, the distance range from the vehicle position is 5m to 50m), the second confidence interval AREA② (for example, the distance range from the vehicle position is 50m to 100m), and the third confidence interval AREA③ (for example, the distance from the vehicle position is more than 100m). For these three confidence intervals, the confidence of the first confidence interval AREA① is greater than the confidence of the second confidence interval AREA②, and the confidence of the second confidence interval AREA② is greater than the confidence of the third confidence interval AREA③.

[0070] It can be understood that as the vehicle moves forward, each confidence interval will move forward.

[0071] The second confidence level is designed with full consideration of the detection characteristics of the radar sensor, that is, the detection capability for objects that are closer is higher than that for objects that are farther away. In this way, setting the detection results of detection points that are closer to the target to a higher confidence level and setting the detection results of detection points that are farther away to a lower confidence level can improve the objectivity and accuracy of the calculation results.

[0072] In one embodiment of calculating the third confidence level, the calculation module 13 matches the boundary of the detected lane with the boundary of each candidate lane, and determines the third confidence level based on the degree of matching. The higher the degree of matching, the higher the third confidence level.

[0073] See also Figure 4 , P_11~P_1n and P_21~P_2n (see Figure 5The "☆" icon in the figure represents the radar detection point. The dotted line L_x1 connecting the radar detection points P_11 to P_1n and the dotted line L_x2 connecting the radar detection points P_21 to P_2n are the two boundaries of the detection lane (radar detection lines). L11 and L12 are the two boundaries of the first candidate lane. L21 and L22 are the two boundaries of the second candidate lane. When matching the boundary of the candidate lane with the boundary of the detection lane, the left boundary of the detection lane is matched with the left boundary of the candidate lane, and the right boundary of the detection lane is matched with the right boundary of the candidate lane, with the vehicle's forward direction as a reference.

[0074] For example, the left boundary L_x1 of the detection lane is matched with the left boundary L21 of the second candidate lane, the shortest distance between each detection point on L_x1 and L21 is calculated, and all the shortest distances are summed. The smaller the sum of the shortest distances, the higher the matching degree. Similarly, the matching degree between the right boundary L_x2 of the detection lane and the right boundary L22 of the second candidate lane can be calculated. In this way, the third confidence of the second candidate lane can be obtained based on the matching degree of the left and right boundaries. In the same way, the third confidence of other candidate lanes can be calculated.

[0075] In this embodiment, the determination of the third confidence level may also take into account the second confidence level, that is, the third confidence level is determined based on the shortest distance between the detection point and the candidate lane boundary and the second confidence level. For example, the third confidence level may be calculated for each confidence interval. The shortest distance between the detection point and the candidate lane boundary is multiplied by the weight of the confidence interval in which the detection point is located. The farther the confidence interval is from the vehicle, the lower the weight. That is, the interval closest to the vehicle (e.g., the first confidence interval) has the highest weight, and the interval farthest from the vehicle (e.g., the third confidence interval) has the lowest weight.

[0076] It is understood that the embodiments of the present invention are intended to fuse map data with radar data, thereby eliminating the problem of lane misjudgment, so as to provide accurate geo-fence event decisions for vehicles. The embodiments of the present invention may include any matching algorithm, confidence algorithm, and combination calculation method of various types of confidence that are suitable for achieving the above purpose, and are not limited to the above examples.

[0077] Next, the decision module 14 determines whether the determined lane decision (box 260) triggers a geo-fence event for the vehicle, that is, based on the vehicle's current lane (the lane it is currently traveling in), whether the vehicle enters the geo-fence and starts automatic driving; or based on the vehicle's current lane (the lane it is currently traveling in), whether the vehicle leaves the geo-fence and ends automatic driving.

[0078] During this process, module 14 may also consider some reference factors to assist in making decisions.

[0079] In one embodiment, module 14 may consider the above map-assisted information to assist in making decisions. For example, although a decision can be made for the vehicle to enter the geofence based on the determined lane, according to the requirements of the geofence and the autonomous driving system, the geofence event is triggered and the autonomous driving is performed only on the highway, and according to the map-assisted information, it can be known that the current road is a provincial road (i.e., a non-highway road), and module 14 makes a decision not to trigger the geofence event based on this.

[0080] In another embodiment, module 14 may consider the above radar auxiliary information to assist in making decisions. For example, although a decision can be made to enter a geofence based on the determined lane, according to the requirements of the geofence and the autonomous driving system, the geofence event is triggered and the autonomous driving is performed only on the highway. According to the radar auxiliary information, it can be known that the average speed on the current road is much lower than the speed on the highway. At this time, mode 14 can take measures such as reducing the calculated confidence or recalculating the confidence. In this embodiment, the radar detection of dynamic vehicles is integrated with the lane speed limit to further improve safety and reliability.

[0081] It is understandable that the module 14 may consider one or both of the above-mentioned map auxiliary information and radar auxiliary information in the decision-making process.

[0082] The following describes the application of the embodiments of the present invention in typical scenarios.

[0083] Multiple parallel roads may include lanes with different types of restrictions. It may be difficult to distinguish which lane of the multiple parallel lanes the vehicle is in based on positioning based on vehicle navigation signals. For application scenarios with parallel roads, the geo-fenced area for performing autonomous driving may only include one of the lanes, rather than all of the lanes. According to an embodiment of the present invention, the radar is capable of detecting guardrail information on both sides of the road, that is, the boundaries of the above-mentioned detection lanes. By integrating the guardrail information detected by the radar with the lane line information in the map data, it is possible to determine which lane of the multiple parallel lanes the vehicle is in.

[0084] For the application scenario of elevated roads, the geo-fenced area where autonomous driving may be performed only includes the lanes above the elevated road, but not the lanes below the elevated road. It is difficult to distinguish whether the vehicle is in the lane above the elevated road or the lane below the elevated road based on the positioning of the vehicle navigation signal. Considering that the roads above the elevated road usually have guardrails on both sides, while the lanes below the elevated road usually have guardrails only on the right side (with reference to the vehicle's forward direction). According to an embodiment of the present invention, the radar is capable of detecting the guardrail information at the boundary of the road. By integrating the guardrail information detected by the radar with the elevated road information in the map data, it is possible to determine whether the vehicle is in the lane above the elevated road or the lane below the elevated road.

[0085] It can be seen that the embodiments of the present invention do not involve accurate positioning of the vehicle position, but are intended to eliminate misjudgment of the vehicle's driving lane. Therefore, the embodiments of the present invention do not require a large amount of positioning map data, but only obtain the map data of relevant roads and lanes in the high-precision map, and then fuse the map data with the radar data to solve the problem of lane misjudgment, thereby providing accurate geo-fence event decisions.

[0086] Figure 5 A method 500 for determining a vehicle's geo-fence event according to a feasible embodiment of the present invention is shown. The method 500 can be executed by the control unit 10 and / or the system 100 as described above, and therefore, the above description of the control unit 10 and the system 100 also applies here.

[0087] See also Figure 5 In step 510, first road information is generated based on the received map data and vehicle position data, and the first road information at least includes one or more candidate lanes based on the vehicle position.

[0088] In step 520 , second road information is generated based on the received radar data, where the second road information at least includes a detected lane based on the radar data.

[0089] In step 530 , the first road information and the second road information are fused and calculated to obtain the confidence of each candidate lane, and the lane in which the vehicle is located is determined based on the calculated confidence.

[0090] In step 540 , a geo-fence event is triggered based on the determined lane decision, where the geo-fence event includes an event of the vehicle entering the geo-fence and an event of the vehicle leaving the geo-fence.

[0091] The above steps 510 and 520 may be performed in any order or simultaneously.

[0092] It is to be understood that all operations in the method described above are merely exemplary, and the present invention is not limited to any operation in the method or the order of these operations, but should cover all other equivalent transformations under the same or similar concept.

[0093] The present invention also provides a machine-readable storage medium storing executable instructions, which, when executed, enables the machine to perform the method 500 described above.

[0094] It is understood that the control unit and modules described above can be implemented in various ways, for example, as hardware, software, or a combination thereof.

[0095] Control unit may include one or more processors. These processors may be implemented using electronic hardware, computer software or any combination thereof. Whether these processors are implemented as hardware or software will depend on specific application and the overall design constraints imposed on the system. As an example, the processor provided in the present invention, any part of the processor or any combination of the processor may be implemented as a microprocessor, a microcontroller, a digital signal processor (DSP), a field programmable gate array (FPGA), a programmable logic device (PLD), a state machine, a gate logic, a discrete hardware circuit and other suitable processing components configured to perform the various functions described in the present invention. The function of the processor provided in the present invention, any part of the processor or any combination of the processor may be implemented as software performed by a microprocessor, a microcontroller, a DSP or other suitable platforms.

[0096] Software can be broadly considered to represent instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, running threads, processes, functions, etc. Software can reside in a computer-readable medium. A computer-readable medium can include, for example, a memory, which can be, for example, a magnetic storage device (e.g., a hard disk, a floppy disk, a magnetic stripe), an optical disk, a smart card, a flash memory device, a random access memory (RAM), a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), a register, or a removable disk. Although the memory is shown as being separated from the processor in the various aspects provided by the present invention, the memory can also be located inside the processor (e.g., a cache or a register).

[0097] The above description is provided to enable any person skilled in the art to implement the various aspects described herein. Various modifications to these aspects are apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects. Therefore, the claims are not intended to be limited to the aspects shown herein. All structural and functional equivalents of elements of the various aspects described herein that are known or to be known to those skilled in the art will be expressly included herein by reference and are intended to be covered by the claims.

Claims

1. A control unit for deciding a geo-fence event of a vehicle, comprising: A first road information generating module, configured to generate first road information based on the received map data and the vehicle position data, which at least includes one or more candidate lanes based on the vehicle position; A second road information generating module, which is configured to generate second road information based on the received radar data, which at least includes a detected lane based on the radar data; a calculation module configured to perform a fusion calculation on the first road information and the second road information to obtain a confidence level of each candidate lane, and determine the lane in which the vehicle is located based on the calculated confidence level; as well as A decision module is configured to trigger a geo-fence event based on the determined lane decision, wherein the geo-fence event includes an event of the vehicle entering the geo-fence and an event of the vehicle leaving the geo-fence.

2. The control unit according to claim 1, wherein: When the candidate lanes include only one lane, the calculation module determines the one lane as the vehicle in which the vehicle is located when the calculated confidence of the one lane is greater than or equal to the confidence threshold; or When the candidate lane includes multiple lanes, the calculation module determines the lane where the vehicle is located by the lane whose calculated confidence is the highest and is greater than or equal to the confidence threshold.

3. The control unit according to claim 1, wherein: The confidence of each candidate lane is calculated by one or more of the following: A first confidence level, which is associated with the shortest distance between the vehicle position and the centerline of each candidate lane; a second confidence level associated with the vehicle position and the distance between each radar detection point; as well as; A third confidence level is associated with the degree of match between the boundary of the detected lane and the boundaries of each candidate lane.

4. The control unit according to claim 3, wherein: The confidence level is determined by the following formula: C i =W1×C i1 + W2×C i2 + W3×C i3 , Among them, C i represents the confidence of the i-th candidate lane; C i1 represents the first confidence of the i-th candidate lane, W1 represents C i1 The weight of C i2 represents the second confidence of the i-th candidate lane, W2 represents C i2 The weight of C i3 represents the third confidence of the i-th candidate lane, and W3 represents C i3 The weight of Among them, the weight C i1~ C i3 It is adjustable.

5. The control unit according to claim 3, wherein: The calculation module is configured to set a plurality of confidence intervals along the vehicle heading, and assign a confidence level to each confidence interval according to the distance between the radar detection point in each confidence interval and the vehicle as the second confidence level.

6. The control unit according to claim 3, wherein: The calculation module is configured to determine the third confidence level based on a shortest distance between a detection point on a detection lane boundary and a boundary of a candidate lane.

7. The control unit according to claim 6, wherein: The third confidence level is determined based on both (1) the shortest distance between a detection point on a detection lane boundary and a boundary of a candidate lane; and (2) the second confidence level.

8. The control unit according to claim 1, wherein: The first road information also includes map assistance information, and the decision module considers the map assistance information when making the decision; and The map auxiliary information includes one or more of the following: road grade, road type, lane type, and boundary type and shape of candidate lanes.

9. The control unit according to claim 1, wherein: The second road information further includes radar assistance information, and the decision module considers the radar assistance information when making the decision; and The radar auxiliary information includes the average vehicle speed in the lane and surrounding lanes.

10. The control unit according to claim 1, wherein: The first road information generation module is configured as follows: One or more candidate lanes within the predetermined search range and information related to the one or more candidate lanes are extracted from the map data according to the vehicle position and the predetermined search range, and the boundary of each candidate lane is determined.

11. The control unit according to claim 10, wherein: If no candidate lanes can be found within the predetermined search range, the predetermined search range is increased.

12. The control unit of claim 1, wherein: The second road information generating module is configured to fit a plurality of radar detection points contained in the radar data into a linear pattern as a boundary of the detection lane.

13. The control unit of claim 1, wherein: The control unit is communicatively connected to an autonomous driving control unit of the vehicle to transmit decisions regarding geo-fence events to the autonomous driving control unit.

14. A system for determining a geo-fence event for a vehicle, comprising: a networked device configured to receive map data for autonomous driving; a positioning device configured to receive a vehicle positioning signal and generate vehicle position data; a radar sensor configured to capture road information of a road ahead of the vehicle and generate radar data; as well as The control unit as described in any one of claims 1-13 is configured to perform fusion calculations on map data, vehicle position data and radar data, and decide whether to trigger a geo-fence event for the vehicle, wherein the geo-fence event includes an event of the vehicle entering a geo-fence and an event of the vehicle leaving a geo-fence.

15. The system of claim 14, wherein: The radar sensor includes one or more of the following: millimeter wave radar, laser radar, ultrasonic radar; Among them, the radar sensor is a long-range millimeter-wave radar installed at the front of the vehicle.

16. A method for determining a geo-fence event of a vehicle, the method being executed by a control unit according to any one of claims 1 to 13 and / or a system according to any one of claims 14 to 15, the method comprising: generating first road information based on the received map data and the vehicle position data, the first road information comprising at least one or more candidate lanes based on the vehicle position; generating second road information based on the received radar data, the second road information comprising at least a detected lane based on the radar data; Performing a fusion calculation on the first road information and the second road information to obtain a confidence level of each candidate lane, and determining the lane in which the vehicle is located based on the calculated confidence level; as well as Based on whether the determined lane decision triggers a geo-fence event, the geo-fence event includes an event of the vehicle entering the geo-fence and an event of the vehicle leaving the geo-fence.

17. A machine-readable medium having machine-readable instructions stored thereon, which, when executed, cause the machine to execute the method of claim 16.

Images