Reliable anchor points for real-time mapping with mobile devices

Abstract

The present invention is a method, system, and computer program product for map data collection. When the mobile data collection device enters a geographic anchor point creation area, an anchor point creation mode is initiated on the mobile data collection device. In the anchor point creation mode, the mobile data collection device collects and distributes raw mapping data to a remote computing device. A high accuracy map of the anchor point creation area is created by combining the received raw data with mapping data from other data sources. After the map is created, when the mobile data collection device enters the anchor point creation area, information from the map is used in a data capture mode to refine a determination of the position and orientation of the mobile data collection device and the position and orientation of one or more features detected by the mobile data collection device.

Description

[Technical field]

[0001] The present invention relates to mapping, and more particularly to collecting data to create a digital map of an environment. [Background technology]

[0002] Mapping generally refers to the process of processing data collected by one or more sensors to create a high definition (HD) digital map of an environment such as a city (also commonly referred to as a "digital twin"). The data collection process typically uses a set of different sensors (optical, radar, sonar, etc.). These sensors can be, for example, one-dimensional (single beam) or two-dimensional (sweeping) laser range finders, three-dimensional HD LIDAR, three-dimensional flash LIDAR, two-dimensional or three-dimensional sonar sensors, one or more two-dimensional cameras, etc. The sensors are usually mounted on dedicated vehicles, and in typical operations, a small fleet of such vehicles travels through different parts of the area to be mapped (e.g. a city) to collect map data. The collected data is processed offline by powerful computers to create the HD digital map. Summary of the Invention [Problem to be solved by the invention]

[0003] Although data collected by dedicated vehicles and sensors meets the requirements for map construction, such data collection methods require specialized and expensive equipment for both data collection and subsequent processing of the collected data. As a result, such traditional map construction techniques are not scalable. Furthermore, it is often difficult to cover a large area such as a city with only a few vehicles, and it may take weeks or months for a vehicle to travel a place or route again. As a result, maps created based on the collected data may be out of date and may not reflect recent changes in the environment, which may have a significant impact on other systems that rely on the map to perform their functions (e.g., navigation applications). Therefore, improvements in mapping technology are needed. [Means for solving the problem]

[0004] According to a first aspect, the present invention relates to a method for collecting map data, wherein an anchor point creation mode is activated on a mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area. In the anchor point creation mode, the mobile data collection device collects and delivers raw mapping data to a remote computing device. A high accuracy map of the anchor point creation area is created at least in part by the remote computing device by combining the raw data received by the remote computing device with mapping data from other data sources. After the high accuracy map is created, when the mobile data collection device enters the anchor point creation area, a data capture mode uses information from the high accuracy map to refine a determination of one or more of the location of the mobile data collection device, the orientation of the mobile data collection device, the location of one or more features detected by the mobile data collection device, and the orientation of one or more features detected by the mobile data collection device.

[0005] Embodiments of the invention include one or more of the following features: The mapping data from other data sources is one or more of maps, surveys (i.e., data collected by surveyors using surveying equipment), and aerial photographs. The remote computing device may be a cloud computing device. The mobile data collection device may be configured to determine the location of the mobile data collection device and the locations of features detected by the mobile data collection device using GNSS sensors, inertial sensors, magnetometers, and camera sensors.

[0006] The method may be configured to use information from the high definition map to refine a previously determined location of one or more features within the anchor point creation region or a location of a feature subsequently detected outside the anchor point creation region. The method may be configured to determine a quality score for one or more features of interest based on the aggregated mapping data, evaluate the quality score of the one or more features of interest against a threshold, identify whether the feature of interest meets the anchor point criteria, and, if the feature of interest meets the anchor point criteria, designate the feature of interest that meets the anchor point criteria as an anchor point within the anchor point creation region. The feature of interest may be one or more of a road sign, a street light, a garbage can, a bench, a bus stop, and a street fixture.

[0007] The mobile data collection device may be configured to switch from the anchor point creation mode to the data capture mode when the mobile data collection device leaves the anchor point creation area, which may be defined by a user. Upon determining that a sufficiently detailed high precision map has been created for the anchor point creation area, the mobile data collection device may be configured to operate in the data capture mode both within and outside the anchor point creation area.

[0008] The mobile data collection device may include a cell phone camera or a camera mounted on a vehicle and a neural network trained to detect one or more categories of physical features in images captured by the cell phone camera or the mounted camera. The mobile data collection device may further include a wireless communication device configured to transmit and receive collected data to and from a cloud-based platform. The mobile data collection device may be a ground-based mobile data collection device. The mobile data collection device may be an autonomous vehicle.

[0009] According to a second aspect, the invention relates to a system for collecting map data, the system comprising a memory and a processor, the memory including instructions which, when executed by the processor, cause the processor to carry out a method, the method comprising: activating an anchor point creation mode on the mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area, whereby the mobile data collection device collects raw mapping data and delivers it to a remote computing device; generating, at least in part by the remote computing device, a high resolution map of the anchor point creation area by combining the raw data received by the remote computing device with mapping data from other data sources; and when the mobile data collection device enters the anchor point creation region after the high definition map is created, in a data capture mode, using information from the high definition map to refine a determination of one or more of a location of the mobile data collection device, an orientation of the mobile data collection device, a location of one or more features detected by the mobile data collection device, and an orientation of one or more features detected by the mobile data collection device. This includes:

[0010] According to a third aspect, the present invention relates to a computer program product for collecting map data, the computer program product comprising a computer readable storage medium having program instructions, the computer readable storage medium being not itself a transitory signal, the program instructions being executable by a processor to carry out a method, the method comprising: activating an anchor point creation mode on the mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area, whereby the mobile data collection device collects raw mapping data and delivers it to a remote computing device; generating, at least in part by the remote computing device, a high resolution map of the anchor point creation area by combining the raw data received by the remote computing device with mapping data from other data sources; and when the mobile data collection device enters the anchor point creation region after the high definition map is created, in a data capture mode, using information from the high definition map to refine a determination of one or more of a location of the mobile data collection device, an orientation of the mobile data collection device, a location of one or more features detected by the mobile data collection device, and an orientation of one or more features detected by the mobile data collection device. This includes: [Brief description of the drawings]

[0011] [Figure 1] 1 is a flow chart illustrating a method for determining anchor points according to one embodiment. [Diagram 2] FIG. 2 is a schematic diagram illustrating a road where a mobile device is entering an anchor point creation area, according to one embodiment. [Diagram 3] 1 is a schematic diagram illustrating a mobile device and a road located within an anchor point creation area according to one embodiment. [Figure 4]1 is a schematic diagram illustrating a mobile device and a road exiting an anchor point creation area according to one embodiment. [Figure 5A] FIG. 2 is a schematic diagram illustrating a mobile device operating in a data collection mode, according to one embodiment. [Figure 5B] FIG. 2 is a schematic diagram illustrating a mobile device operating in a data collection mode, according to one embodiment. [Figure 6] FIG. 1 is a schematic diagram illustrating a two road intersection and multiple mobile devices collecting anchor point data, according to one embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012]

[0013] One or more embodiments of the invention are set forth in detail in the accompanying drawings and the description below. Other features and advantages of the invention will become apparent from the detailed description and drawings, and from the claims.

[0013] Like reference characters indicate like elements throughout the drawings.

[0014] As mentioned above, an objective of embodiments of the present invention is to reduce the complexity and costs associated with creating HD digital maps of various environments, such as cities or parts of cities, particularly with respect to the data collection step. Another goal is to obtain more frequent (and therefore more up-to-date) map data compared to existing data collection methods, which can be used to update the digital maps more frequently and improve their accuracy, thereby improving other applications that use digital maps.

[0015] As described in more detail below, embodiments of the present invention use a standard sensor setup on a mobile data collection device (hereafter referred to as a mobile device), such as a mobile phone or Internet of Things (IoT) device, to collect data and create an HD map using the data, i.e., a camera (although multiple cameras may be used for devices with multiple cameras), an inertial measurement unit (IMU), sensors associated with a global navigation satellite system (e.g., GPS, GLONASS, Baidu, etc.), or a combination thereof.

[0016] The mobile device is mounted, either permanently or temporarily, on a vehicle that moves autonomously or under the control of an operator on the ground or in the air. Examples of vehicles include automobiles, aircraft, bicycles, drones, or other manned or unmanned vehicles. The mobile device also includes, or in some embodiments is connected to, a computing device or processor and a data link, such as a cellular 4G or 5G connection. Configured in this manner, the mobile device can perform pre-processing of collected image and location data before transmitting the data over the data link to a cloud-based platform for further processing. Performing pre-processing can significantly reduce the amount of data being transmitted from the mobile device to the cloud-based platform, making the transmission more efficient. The mobile device can also receive settings and software, such as image processing or neural network software, tailored to the particular type of mobile device (e.g., connected car vs. smartphone) or the type of data the mobile device is expected to collect and pre-process, allowing the mobile device to operate in different "modes," as described in more detail below.

[0017] To enrich data collection, embodiments of the present invention intensively utilize the sensor configuration and computational power of the mobile device to automatically create a sparse set of highly accurate reference points (also referred to herein as "anchor points") in a short period of time. Such mobile device operation is hereinafter referred to as "anchor point creation mode". In short, in the anchor point creation mode, data is streamed from the mobile device to a cloud-based platform (either a private cloud-based platform, a public cloud-based platform, or a hybrid of these), the collected data is processed, and HD maps and anchor points are generated. The generated anchor points can then be used for map generation and updating by other mobile devices, while the mobile device can be used in a more sustainable manner, hereinafter referred to as "data capture mode", which can reduce battery drain and the amount of data sent to the cloud-based platform.

[0018] In particular, as described in detail below, embodiments of the present invention provide a mechanism for automatically triggering an anchor point creation mode on a mobile device, a technique for generating anchor points based on sensor data available on the mobile device, and a technique for utilizing anchor points to update a map generated using one or more mobile devices operating in a data capture mode. Various embodiments of the present invention will now be described by way of example with reference to the drawings. Note that the examples given below generally refer to a single mobile device, a single road, a single anchor point creation area, etc., but this is only for ease of understanding the operation of the embodiments of the present invention. In practical use scenarios, there will generally be many anchor point creation areas, mobile devices, and the actual method will generally be continuous with few interruptions.

[0019] Figure 1 is a flow chart illustrating a method 100 for creating and using HD maps, according to one embodiment of the present invention. As can be seen from Figure 1, the method begins by a user defining a geographical area, e.g., a portion of a city for which an HD map is to be created (step 102). This area is called an anchor point creation area. The designation of the anchor point creation area can be done in a semi-automated manner, where experts determine the most suitable locations for anchor point creation based on analysis of factors such as historical data, e.g., data collected in the past year, cross-correlation with areas with similar characteristics, feedback from users, and experience of engineers.

[0020] FIG. 2 shows a schematic example of an anchor point creation region 200. Although the anchor point creation region 200 is shown in the shape of a circle, it can be of any shape, as will be appreciated by those skilled in the art. Typically, the anchor point creation region is defined by a user, for example by outlining the anchor point creation region 200 on a digital map of a city or other area of ​​interest. However, various degrees of automation may be applied in defining the anchor point creation region 200. For example, in one embodiment, the anchor points are distributed evenly throughout a space such as a city, while other embodiments create anchor points based on complexity metrics (e.g., number of tall buildings, density of buildings, proximity to a city center, etc.). Furthermore, in another embodiment, anchor points are automatically created based on detected feature location errors (e.g., large discrepancies between feature locations reported by different agents), as will be described in more detail below. This may indicate poor performance and provide information about locations that could be enhanced by anchor points. However, those skilled in the art may envision further criteria for creating the anchor point creation region 200.

[0021] In one embodiment, anchor point creation area 200 defines a so-called geofencing area, which allows for control over various aspects of a mobile device's entry into anchor point creation area 200. For example, Figure 2 shows a mobile device 202 traveling on a road 204 that passes through anchor point creation area 200. As can be seen in Figure 2, mobile device 202 is located just outside anchor point creation area 200. Figure 2 also shows a road sign 206 located inside anchor point creation area 200.

[0022] The next step in the method 100 is to determine whether the mobile device 202 is located within the anchor point creation region 200 (step 104). This is shown diagrammatically in Figure 3. This can be done by simply comparing the location of the mobile device 202 against the anchor point creation region 200 (i.e., the geofenced region), for example using GPS technology as is conventional in the art.

[0023] As shown in FIG. 3, when the mobile device 202 is determined to be within the anchor point creation region 200 in step 104, the mobile device 202 automatically switches to an anchor point creation mode (step 106). In the anchor point creation mode, the mobile device 202 starts computationally intensive data collection and processing. Compared to when the mobile device operates in a normal data capture mode, a high rate of image and sensor data is acquired, the number of data samples 302 collected is significantly increased, and the location of the feature of interest is estimated with a higher accuracy than is possible in the data capture mode. The collected data is "delivered" to a cloud-based platform, where the data is further processed and an HD map is generated. Typically, data collection is highly automated and can use a variety of techniques, such as 3D positioning, 3D mapping, 3D localization, spatial deep learning, Visual-inertial Simultaneous Localization and Mapping (SLAM), visual odometry, object detection, tracking, etc., either individually or in combination.

[0024] Depending on the processing power of the mobile device 202, various levels of pre-processing of the collected data may be performed on the mobile device 202. This can reduce the amount of data that is ultimately sent from the mobile device 202 to the cloud-based platform. In general, high-load data collection and processing incurs costs in terms of battery drain on the mobile device 202 when processing and sending the collected data to the cloud-based platform. If the size of the anchor point creation area 200 is too large, there is a risk that the mobile device 202 will run out of battery, so generally, some constraints will be placed on the size of the anchor point creation area 200.

[0025] As described above, anchor points are high-precision reference points that are used as references by mobile devices 202 operating in data capture mode to improve data collection for new or existing maps. Often, anchor points are road signs, such as no parking signs 206, located within the anchor point creation area 200. This is because road signs are distinctive and unlikely to change over time. However, in various embodiments described herein, the anchor point creation area 200 is not limited to only road signs 206. Essentially, any salient feature that is highly visible (i.e., easily detectable by computer vision in any weather condition) and unlikely to change over time can be used as an anchor point, such as a corner of a building, a permanent artwork, a bench, a landscaping feature, etc. A salient feature may not necessarily be "meaningful" from a human perspective, but may be important from a computer vision perspective. Anchor points may be assigned a "quality score" that indicates how reliable the anchor point is. In one embodiment, the score is compared to a reliability threshold to indicate whether the anchor point is reliable or not. In other embodiments, once the quality score exceeds a minimum confidence threshold, one feature may be deemed better than others, taking into account variables such as the environment (e.g., weather, lighting, time of day), orientation and trajectory (e.g., approaching an anchor point from a particular direction), among others.

[0026] In an embodiment, to further improve data collection, data from multiple mobile devices 202 operating in a data anchor point creation mode are combined as shown in FIG. 6. In the example shown in FIG. 6, the same road sign 206 is detected by three mobile devices 202a, 202b, and 202c. The mobile devices 202a, 202b, and 202c are all located within the anchor point creation area 200 and are also operating in an anchor point creation mode. The three different mobile devices detect the road sign 206 from three different directions and each reports the location of the road sign 206 to the cloud-based platform. This allows the cloud-based platform to combine the information received from the mobile devices 202a, 202b, and 202c to determine the location of the road sign 206 with high accuracy. As a result, the road sign 206 is considered a reliable anchor point that other mobile devices can rely on for the accuracy of their location determination. 6, it appears that the three mobile devices 202a, 202b, 202c detect the road sign 206 simultaneously, but the detection does not have to be simultaneous. In fact, it may be advantageous for the three mobile devices 202a, 202b, 202c to detect the road sign 206 in different conditions (e.g., with respect to light, weather, etc.).

[0027] In one embodiment, once a sufficient number of anchor points 206 of acceptable quality have been collected within the anchor point creation area 200, the mobile device 202 entering the anchor point creation area 200 continues to operate in data collection mode rather than activating the anchor point mode. The criteria for determining whether a sufficient number of anchor points 206 of acceptable quality have been collected may be defined by the user of the method and is typically determined based on various conditions specific to the particular situation at hand, such as the number of times the mobile device 202 has passed through the anchor point creation area 200, the number of times the mobile device 202 has passed through the anchor point creation area 200 from different directions, the discrepancy between multiple measurements of the anchor point's location and orientation, comparison (validation) with a known source of high accuracy but outdated information, etc.

[0028] 1, as shown in FIG 4, when the mobile device 202 is not located within the anchor point creation area 200, either because the mobile device 202 has not entered or has left the anchor point creation area 200, the mobile device 202 operates in the data capture mode (step 108). Switching from the anchor point creation mode to the data capture mode may occur automatically when the mobile device 202 leaves the geofenced anchor point creation area 200.

[0029] 5A and 5B conceptually illustrate a mobile device 202 moving along a road 204 and operating in a data capture mode. The mobile device 202 may be a mobile device previously operating in an anchor point creation mode, as described above, or it may be an entirely different mobile device 202. Either way, the same principles apply. As shown in FIG. 5A, the mobile device 202 moves along a road 204 and locates a previously determined anchor point 206. The mobile device 202 may determine an error between the estimated location of the anchor point 206 and the actual location of the anchor point 206 determined when the anchor point 206 was created. This error is typically caused by bias or noise in the mobile device 202's sensors, but may also be caused by other factors such as lighting (e.g., low sun position, reflections on the surroundings) and vibration (e.g., when the mounted vehicle vibrates due to engine or poor road conditions). The mobile device 202 may use the anchor point 206 to correct its own estimated location. As the mobile device 202 travels along the roads 204 and detects new features (e.g., road signs 502, 504) that are not present in the map of the area, the locations of these new features 502, 504 are determined with greater accuracy (i.e., with respect to the anchor points 206) as compared to when the mobile device 202's self-determined location was not corrected, as shown diagrammatically in FIG. 5B. In one embodiment, the approximate locations of the new features 502, 504 are determined by traditional methods, e.g., GPS, and then corrected using the precise location of the anchor points 206. Any detected new features 502, 504 are sent to a cloud-based platform and the HD map is updated to include the new features 502, 504.

[0030] However, the anchor points 206 may be used to improve the determination of the position and orientation of the mobile device 202 outside the anchor point creation region 200. For example, a small error in the position or orientation of the mobile device 202 (i.e. roll, pitch, yaw (Euler angles) or a certain quaternion for a certain datum) may grow as the mobile device 202 moves away from the point where the position determination error occurred. As a result, the final position of the mobile device 202 as determined by the mobile device 202 itself may be significantly different from its actual position, and measurements of newly detected features 502, 504 along the way may be assigned incorrect positions. Thus, by using reliable anchor points, the actual position of the mobile device 202 can be compared with the measured positions of the mobile device 202 and necessary corrections can be made, thus preventing or at least significantly reducing these types of errors. As described above, good position determination of the mobile device 202 in the vicinity of an anchor point will result in good position determination of the mobile device 202 (and thus discovery of new features) even when the mobile device 202 moves away from the anchor point and the general anchor point creation area.

[0031] However, at some point, the error between the estimated location of the mobile device itself and its actual location may become too large, and it may be desirable to create another anchor point creation region 200. Whether such a need exists and whether it should be addressed will depend on the particular situation at hand, and can be easily determined by one of ordinary skill in the art. In making this determination, various measurements may be taken into account, such as the time since the last "recalibration" of the mobile device's location, the estimated error received from the satellites involved in the global positioning measurements, etc. Thus, for example, a city may have many separate anchor point creation regions, with more anchor points in areas that are difficult to cover by satellites, such as areas with many tall buildings, trees, etc.

[0032] Thus, in contrast to existing solutions, embodiments of the present invention do not rely on existing maps or existing anchors, which are created using dedicated vehicles with expensive sensor setups. In embodiments of the present invention, anchor points are generated in a semi-automated or fully automated process, where the only possible manual input is to define one or more anchor point creation areas, for example in a city for which the creation of anchor points is to be performed. As a result, more up-to-date and accurate HD maps can be created.

[0033] As described above, such anchor points allow for the generation of highly accurate HD maps of cities and their individual regions, which can be used for a wide range of purposes. HD maps may be generated using conventional geographic maps in electronic format, similar to the maps used by various map service providers, such as Google Maps and Apple Maps, by storing a feature dataset in a georeferenced data type format (e.g., GeoJson open standard format) and adding anchor points and other features collected by mobile devices to the map. In some implementations, various types of third-party data are added to the map as needed, such as information about weather, road regulations (e.g., speed limits) on various road segments, an inventory of assets (e.g., traffic signs) installed on the side of the road, temporary changes (e.g., road works), and a map of parking spaces.

[0034] The generated map may be displayed on the display, in a web-based user interface, or in a standalone application implemented locally on the computing device. Typically, the map graphical user interface is configured to allow the user to control zooming in, zooming out, panning, and the like, as well as select which "layers" to display. For example, the user may select to display information such as congestion, road types, speed limits, road construction, and parking availability. After reviewing the displayed information, the user may instruct various systems, devices, and people to take action to mitigate the detected problems.

[0035] Certain or all of the components may be implemented as software executed by a digital signal processor or microprocessor, or as hardware or application specific integrated circuits. Such software may be distributed on computer readable media consisting of computer storage media (or non-transitory media) and communication media (or transitory media). As will be appreciated by those skilled in the art, the term computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information such as computer readable instructions, data structures, program modules, or other data. Computer storage media may include, but is not limited to, RAM, ROM, EEPROM, flash memory, or other memory technology, CD-ROM, digital versatile disks (DVDs), or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage, or other magnetic storage devices, or other media that may be used to store the desired information and that may be accessed by a computer.

[0036] The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowcharts or block diagrams may represent a module, segment, or portion of instructions constituting one or more executable instructions for implementing a specified logical function. In alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two successive blocks may in fact be executed substantially simultaneously or may be executed in the reverse order depending on the functionality involved. Also, each block in the block diagrams or flowcharts, and combinations of blocks in the block diagrams or flowcharts, may be implemented by a special purpose hardware-based system or combination of special purpose hardware and computer instructions to perform the specified functions or acts.

[0037] While the embodiments described above use image sensors such as cameras, other types of sensors may be used, such as optical sensors, infrared imaging sensors, ultraviolet imaging sensors, light detection and ranging (LIDAR) sensors, synthetic aperture radar (SAR) sensors, electromagnetic (EM) sensors, acoustic sensors, or other sensors capable of generating one or more images of physical features in the physical environment.

[0038] Although the present invention has been illustrated in an embodiment configured for use in a city or part of a city, it may also be used in different environments such as warehouses, mines, etc. Essentially, any environment in which a digital twin can be created and used for navigation purposes may be considered to be within the scope of the present invention.

[0039] It is obvious to those skilled in the art that the advantages of the present invention shown in the above-mentioned embodiment can be obtained even when various modifications are made to the above-mentioned embodiment. Therefore, the present invention is not limited to the described embodiment, but should be defined only by the appended claims. Furthermore, as is obvious to those skilled in the art, the above-mentioned embodiments may be combined to form a configuration.

Claims

1. A method for collecting map data, comprising: activating an anchor point creation mode on the mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area, whereby the mobile data collection device collects raw mapping data and distributes it to a remote computing device; generating, at least in part by the remote computing device, a high-precision map of the anchor point creation area by combining the raw data received by the remote computing device with mapping data from other data sources; when a mobile data collection device enters the anchor point creation area after the high precision map is created, in a data capture mode, information from the high precision map is used to refine a determination of one or more of the position of the mobile data collection device, the orientation of the mobile data collection device, the position of one or more features detected by the mobile data collection device, and the orientation of one or more features detected by the mobile data collection device; A method characterized by:

2. The mapping data from other data sources is one or more of maps, surveys, and aerial photographs.

2. The method of claim 1 .

3. the remote computing device is a cloud computing device.

2. The method of claim 1 .

4. the mobile data collection device uses one or more of a GNSS sensor, an inertial sensor, a magnetometer, and a camera sensor to determine the location of the mobile data collection device and the locations of the features detected by the mobile data collection device; 2. The method of claim 1 .

5. using information from the high precision map to refine one or more of previously determined locations of one or more features within the anchor point creation region, locations of features subsequently detected outside the anchor point creation region.

2. The method of claim 1 .

6. determining a quality score for one or more features of interest based on the aggregated mapping data; evaluating the quality scores of the one or more features of interest against a threshold to identify whether the features of interest meet the criteria for an anchor point; and if a feature of interest satisfies the anchor point criteria, designating the feature of interest that satisfies the anchor point criteria as an anchor point within the anchor point creation region.

2. The method of claim 1 .

7. the feature of interest is one or more of a road sign, a street light, a trash can, a bench, a bus stop, or a street fixture; 7. The method of claim 6.

8. and switching the mode of the mobile data collection device from the anchor point creation mode to a data capture mode when the mobile data collection device exits the anchor point creation area.

2. The method of claim 1 .

9. The anchor point creation area is defined by a user.

2. The method of claim 1 .

10. and, upon determining that a sufficiently detailed high precision map has been created for the anchor point creation area, operating the mobile data collection device in the data capture mode both within the anchor point creation area and outside the anchor point creation area.

2. The method of claim 1 .

11. the mobile data collection device, Mobile phone cameras or cameras mounted in vehicles; a neural network trained to detect one or more categories of physical features in the mobile phone camera or images captured by the camera; including one or more of:

2. The method of claim 1 .

12. the mobile data collection device further comprising a wireless communication device; the wireless communication device is configured to transmit and receive collected data to and from a cloud-based platform; 2. The method of claim 1 .

13. the mobile data collection device is a ground-based mobile data collection device; 2. The method of claim 1 .

14. the mobile data collection device is an autonomous vehicle.

2. The method of claim 1 .

15. 1. A system for collecting map data, comprising: Memory and a processor; Equipped with the memory contains instructions that, when executed by the processor, cause the processor to perform a method; The method comprises: activating an anchor point creation mode on the mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area, whereby the mobile data collection device collects raw mapping data and distributes it to a remote computing device; generating, at least in part by the remote computing device, a high-precision map of the anchor point creation area by combining the raw data received by the remote computing device with mapping data from other data sources; when a mobile data collection device enters the anchor point creation area after the high precision map is created, in a data capture mode, information from the high precision map is used to refine a determination of one or more of the position of the mobile data collection device, the orientation of the mobile data collection device, the position of one or more features detected by the mobile data collection device, and the orientation of one or more features detected by the mobile data collection device; Including, A system characterized by:

16. 1. A computer program product for collecting map data, comprising: a computer-readable storage medium having program instructions; The computer-readable storage medium is not itself a transitory signal, the program instructions are executable by a processor to perform a method; The method comprises: activating an anchor point creation mode on the mobile data collection device when the mobile data collection device enters a predetermined geographic anchor point creation area, whereby the mobile data collection device collects raw mapping data and distributes it to a remote computing device; generating, at least in part by the remote computing device, a high-precision map of the anchor point creation area by combining the raw data received by the remote computing device with mapping data from other data sources; when a mobile data collection device enters the anchor point creation area after the high precision map is created, in a data capture mode, information from the high precision map is used to refine a determination of one or more of the position of the mobile data collection device, the orientation of the mobile data collection device, the position of one or more features detected by the mobile data collection device, and the orientation of one or more features detected by the mobile data collection device; Including, 1. A computer program product comprising:

Images