Assessing a work site for autonomous navigability

Abstract

A work region is traversed by a vehicle equipped with a radio receiver and a location sensor. While traversing the work region, geolocations are repeatedly determined via the location sensor, and radio signal characteristics are repeatedly determined via the radio receiver. The radio signal characteristics correspond to the respective geolocations and pertain to a radio signal used for navigation assist. A map of the radio signal characteristics at the geolocations is stored in a computer readable medium. A navigability map of an autonomous ground maintenance machine in the work region is determined based on the map of the radio signal characteristics.

Description

ASSESSING A WORK SITE FOR AUTONOMOUS NAVIGABILITYRELATED PATENT DOCUMENTS

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 546,674, filed on October 31, 2023, and U.S. Provisional Application No. 63 / 597,780, filed on November 10, 2023, both of which are incorporated herein by reference in their entireties.SUMMARY

[0002] The present disclosure is directed to tools for assessing autonomous navigability of a work site, e.g., for navigation by autonomous machines to perform ground care work. In one embodiment, a method involves traversing a work region with a vehicle. The vehicle is equipped with a radio receiver and a location sensor. While traversing the work region, geolocations are repeatedly determined via the location sensor, and radio signal characteristics are repeatedly determined via the radio receiver. The radio signal characteristics correspond to the respective geolocations and pertain to a radio signal used for navigation assist. A map of the radio signal characteristics at the geolocations is stored in a computer readable medium. Via a computer processor, a navigability map of an autonomous ground maintenance machine in the work region is determined based on the map of the radio signal characteristics.

[0003] In another embodiment, a method involves traversing a work region with a vehicle that is equipped with a radio receiver and a location sensor. While traversing the work region, the method involves repeatedly determining geolocations via the location sensor and determining, via the radio receiver, radio signal characteristics corresponding to the respective geolocations. The radio signal characteristics pertain to a radio signal used for navigation assist. A map of the radio signal characteristics at the geolocations is stored in a computer readable media. Via a computer processor, a navigability map of an autonomous ground maintenance machine in the work region is determined based on the map of the radio signal characteristics. The navigability map is used to plan a work path for the autonomous ground maintenance machine. The work path accounts for regionswhere the autonomous ground maintenance machine cannot rely on the radio signal for autonomous navigation.[00041 Inanother embodiment, a system includes a vehicle comprising a measurement apparatus. The measurement apparatus includes a radio receiver and a location sensor coupled to a system controller. The measurement apparatus is operable to, while the vehicle traverses a work region, repeatedly determine geolocation data via the location sensor and determine, via the radio receiver, radio signal characteristics data corresponding to geolocations of the geolocation data. The radio signal characteristics data pertaining to a radio signal used for navigation assist. The system includes a computing arrangement configured to receive the geolocation data and the radio signal characteristics data from the measurement apparatus. The computing arrangement is further operable to store the radio signal characteristics data and the geolocation data in a computer readable medium. The computing arrangement performs a simulation of an autonomous ground maintenance machine traversing the work region to perform work. The simulation uses the radio signal characteristics data and the geolocation data simulating an effect of radio signals on autonomous navigation of the autonomous grounds maintenance machine in the work region. The computing arrangement further determines a navigability map of the autonomous ground maintenance machine in the work region.

[0005] These and other features and aspects of various embodiments may be understood in view of the following detailed discussion and accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The discussion below makes reference to the following figures, wherein the same reference number may be used to identify the similar / same component in multiple figures. The drawings are not necessarily to scale.

[0007] FIG. 1 is a schematic view of a work site according to various example embodiments;

[0008] FIG. 2 is a picture of a vehicle and data gathering apparatus according to an example embodiment;

[0009] FIG. 3 is a graphical representation of a radio signal characterization map according to example embodiments;

[0010] FIGS. 4A, 4B, and 4C are graphical representations of navigability maps according to example embodiments;

[0011] FIGS. 5A and 5B are representations of autonomous navigation through parts of a work region where navigability is below a threshold according to example embodiments;

[0012] FIG. 6 shows an example of a work scenario involving two or more autonomous work vehicle according to an example embodiment;

[0013] FIG. 7 is a block diagram of a system and apparatuses according to an example embodiment; and

[0014] FIGS. 8 and 9 are flowcharts of methods according to example embodiments.DETAILED DESCRIPTION

[0015] In the following detailed description of illustrative embodiments, reference is made to the accompanying figures of the drawing which form a part hereof. It is to be understood that other equivalent embodiments, which may not be described and / or illustrated herein, are also contemplated.

[0016] The present disclosure relates generally to evaluation and planning for autonomous work by ground care machines, which may be variously referred to herein as ground / grounds care vehicles, ground maintenance machines, ground maintenance vehicles, and the like. Ground care machines, such as lawn and garden machines, are known for performing a variety of tasks. For instance, powered lawn mowers are used by both homeowners and professionals alike to maintain grass areas within a property or yard. The same or different machines may be used for other maintenance on the turf areas (and sometimes away from the turf), performing operations such as debris collection, spraying, dethatching, edging, rolling, towing, snow / ice treatment and removal, marking, etc.

[0017] Embodiments of the present disclosure relate to ground maintenance machines that have autonomous functionality. Generally, autonomous functionality may include operations that can be performed without human input that causes or effects a physical action performed by the machine. One example of autonomous operation is autonomous navigation, where the machine can maneuver around a work region without user input, or with minimal user input (e.g., initial placement and initiating a start command). The machine may also be configured to perform autonomous work while navigating, e.g., rotate a cutting blade.

[0018] Use of autonomous machines can provide significant benefits for those who are tasked with grounds maintenance. At times, ground care work can be repetitive, which is often the type of work that lends itself well to automation. For commercial operators, autonomous machines can provide significant savings in labor costs. Also, autonomous machines may be able perform some tasks better than humans, such as following precise lines when mowing grass. Further, autonomous machines may be able to work continuously throughout the day without requiring breaks and can work under extreme outdoor weather conditions such as excessive heat and humidity, poor air quality, etc., that might be unhealthy for a human worker.

[0019] Deploying one or more autonomous machines can be a significant investment in time and materials, even for a relatively simple home installation. Outdoor spaces can have areas in which even human-operated machines can get stuck, and the training of the machines to navigate the space can take a non-trivial amount of time and effort. Further, the availability of wireless navigation aids such as global navigation satellite system (GNSS) and real time kinematics (RTK) may not be guaranteed over the entire work region. Obstacles such as hills and buildings can block the radio signals upon which such navigation relies, leading to a situation where an autonomous machine may not be able to reliably navigate in some areas using those navigation aids.

[0020] There are ways to mitigate loss of wireless geolocation, such as relying on a different mode of navigation such as vision-based navigation with cameras, LIDAR, and the like. Nonetheless, these additional navigation modes may take additional time for setup, training, and the like and may require more expensive hardware be installed on theautonomous machine. Tn order to gauge the cost and effort needed to implement an autonomous work fleet for a large installation (e.g., a golf course, a park), stakeholders (e.g., landscape contractors, property owners) may want to gather significant information about the specifics of their site related to autonomous navigability before investing in an autonomous system.

[0021] In FIG. 1, a block diagram shows details of system according to an example embodiment. A property 100 includes one or more work regions 102. In this example the property 100 may include a housing community or the like, and the work regions 102 include lawns. A vehicle 104 traverses the work regions 102, in some cases to perform regular ground maintenance such as mowing, fertilizing, aerating, etc. In one scenario, a groundskeeper operates one or more such vehicles 104 and may want to consider replacing or augmenting the manually operated machines with autonomous machines. In order to evaluate the site’s suitability for autonomous work, one or more of the vehicles 104 can be outfitted with measurement apparatus 106.

[0022] The measurement apparatus 106 includes a radio receiver 108 and a location sensor 110. The receiver 108 and location sensor 110 may be part of a common device, e.g., a GNSS receiver, or may be separate. Multiple receivers 108 and multiple location sensors 110 may also be used within the apparatus 106. As will be described in greater detail later, the receiver 108 and location sensor 110 repeatedly provide data 114 while traversing at least one work region 102, and this data may be collected stored locally in a data storage medium 112 (as shown) and / or remotely, e.g., in an Internet-accessible data center. This data 114, which include at least geolocations and radio signal characteristics, can be used to determine a navigability map of an autonomous ground maintenance machine (not shown) in the work region 102.

[0023] In the scenario described above, a worker may navigate the vehicle 104 through the work regions 102 in the normal course of performing work while the data 114 is gathered. This may have advantages in cases where the work targeted for performance by an autonomous machine is the same as that performed by the vehicle 104. For example, in such a scenario, the vehicle 104 will likely cover the entirety of the work region 102 that requires work (e.g., turf area that is mowed) and may do so in various different traversals(e.g., cross-hatch pattern) over that ensures a diversity of different measurements over the same regions and over a long period of time, e.g., multiple work sessions over a season. Further, the vehicle 104 in such a scenario may avoid those regions that do not need to be worked, reducing the data collection burden compared to, for example, a full survey of the boundaries of the entire property 100. Another advantage in using a same-type work vehicle relates to other data that can be collected by the vehicle 104 during work, such as potential stuck areas (e.g., detected due to variations in speed, shock and vibration measurements, slope / tilt measurements) or other factors relevant to performing the work, such as best or acceptable times of day in which to perform the work, turn around patterns used by the operator, energy consumed, etc.

[0024] Because the vehicle 104 is already tasked with doing work in the work regions 102, addition of the measurement apparatus 106 and gathering of the data 114 may require little extra effort on the part of the vehicle operator, worker, supervisor, or other involved parties. The vehicle operator need not have any specialized training to gather the data 114, and the installation and setting up of the apparatus 106 need only be performed once, as well as subsequent removal. The apparatus 106 may include communication means (not shown) to report the data 114 and / or status of the data gathering (e.g., number of data points per hour gathered since last power cycle) to a remote entity so that active monitoring of the apparatus 106, e.g., by the vehicle operator, is not unnecessary.

[0025] Note that it is not required that the vehicle 104 be the same type of grounds work vehicle as the autonomous work vehicle that is the target of the navigability survey (referred to herein as the “targeted autonomous vehicle”). For example, the vehicle 104 may be a mower (as shown in FIG. 2) that collects data during the summer, while the targeted autonomous vehicle may be a leaf collector intended for use in the fall, and / or a snow removal vehicle intended for use in the winter. In the latter case, the vehicle 104 may attempt to gather data outside of turf regions, e.g., along sidewalks, driveways, parking lots, and the like, e.g., assuming the operator is aware that this data is desired and can be obtained during regular work. In some embodiments, the vehicle 104 need not be a work vehicle at all, e.g., it may be a golf cart, bicycle, etc., that traverses the property for any purpose. Similarly, the vehicle which ultimately uses the gathered data could be any typeof data besides a mower, e.g., autonomous sprayer, aerator, roller, blower, groomer, snow remover, leaf collector, etc.[00261 The vehicle 104 is described above as operator controlled, e.g., manually guided, driven by the operator, remote controlled via a line-of-sight wireless controller, etc. In some embodiments, the vehicle 104 may be autonomous. Even if the property 100 has not been surveyed for autonomous navigability, some classes of autonomous vehicles may still attempt to traverse the region without prior training in the region. For example, small and slow vehicles can attempt to map a region with minimal impact to the environment or others while moving through the environment. In another example, an autonomous tracking vehicle with a sophisticated (and likely expensive) navigation system may be usable for purposes of gathering the data 114, the data 114 being targeted for use by a less capable (and less expensive) navigation system to do the actual work on the same or different drive chassis. In other embodiments, the targeted autonomous vehicle (with the same navigation system used for work) itself could gather the data, e.g., in a predetermined or random traversal mode within predefined boundaries.

[0027] The location sensor 110 may include a geolocation sensor utilizing GNSS, RTK, and the like. Since the location measurements used for data gathering need not be as precise as for subsequent work navigation, less precise forms of location sensing may be associated with the gathered data 114, such a cell phone tower triangulation, WiFi hotspot detection, visual navigation, etc. For some forms of data such as visual navigation, the images used to estimate location may not need to be identified and converted to a location at the time the data 114 is gathered. One or more images for each location may be gathered with the radio signal data, and then the image data may be post-processed (e.g., by comparing to an established visual mapping database) to extract the geolocation data, such as latitude, longitude, and elevation. The data 114 may also include confidence levels or other indication of the reliability of the gathered location data, as this confidence data may be useful when processing the gathered data.

[0028] The radio receiver 108 may gather multiple types of radio data, including Received Signal Strength Indicator (RSSI), Signal to Noise Ratio (SNR), and other measures of signal degradation, such as multipath interference. The radio receiver 108 maygather metadata as well related to navigation, such as base station identifiers, number of visible satellites, etc. The receiver 108 is coupled to at least one antenna (e.g., antenna 200 in FIG. 2), and may be coupled to different antennas of different types so that the response of different navigation receiver setups can be estimated. The signals gathered may originate from Global Positioning System (GPS) satellites, RTK base stations / repeaters, cell phone base stations / repeaters, WiFi base stations, etc. In order to reduce the amount of data gathered and processed, a subset of frequency ranges may be targeted for collection. For example, if two cell phone carriers are under consideration, then only known spectra from those carriers may be recorded, even if there are other providers transmitting in the area.

[0029] While traversing the work region 102, the measurement apparatus 106 repeatedly determines geolocations via the location sensor 110 and determines, via the radio receiver 108, radio signal characteristics corresponding to the respective geolocations. The radio signal characteristics pertain to a radio signal used for navigation assist of the target autonomous vehicle or vehicles. The apparatus 106 stores a map of the radio signal characteristics to the geolocations in the computer readable storage medium 112. In computer science, the term “map” is generally meant to refer to a data structure of key-value pairs in which in which each value is associated with a key. In this case, the key is a geolocation (e.g., lat / lon) or other location value (e.g., x-y coordinate) and the value may be a radio characteristic such as RSSI l in data set 114. Note that these data sets can be stored in other data structures besides maps, e.g., multi-map, array, linked list, etc., which can be used to present graphical maps, e.g., on a computer screen or print out.

[0030] Note that the term “geolocation data” does not require the data to be globally referenced, e.g., such as latitude and longitude. Other global or non-global coordinate systems may be used, e.g., Universal Transverse Mercator (UTM), property survey coordinates, etc. Further, an arbitrary coordinate system may be used to define the geometry of the work region, e.g., a two-dimensional x-y coordinate system with an arbitrary origin relative to the earth and / or arbitrary rotation relative to a compass heading. Locations used in an arbitrary coordinate system can be translated to global coordinates as known in the art. Since the radio signals gathered typically emanate from services thatprovide global coordinates, a global geolocation value may be a convenient format in which to store the data. Also note that geolocation data may also include heading data (e.g., degrees offset from north) and / or altitude data (e.g., distance from sea level). Such data is generally not required to build the maps as described herein, but may be used for other purposes, such as to augment simulations and for troubleshooting.

[0031] As collected, the data 114 may not conform to the common computer science definition of map, in which each key is unique. For example, there may be multiple measurements made at a single location (e.g. due to the machine being stationary between two measurements), thus the keys may not be unique. Nonetheless, the collected data can be converted to a map through data reduction techniques known in the art (e.g., average points over time and geometric proximity) and there need be no requirements on the precise form of collected data so long as it at least includes radio characteristic data over different locations, and provides a way to link the characteristics to the locations.

[0032] In FIG. 2, a diagram shows the vehicle 104 with measurement apparatus 106 interacting with an infrastructure computing / data center 202 according to an example embodiment. Generally, a data center 202 includes a large number of computers (e.g., thousands), the resources of which can be dynamically distributed between an even larger number of end users. The apparatus 106 may store the map of the radio signal characteristics locally, e.g., on a flash memory drive and / or upload the map data via a network interface 204, such as a cellular data interface. The network interface 204 is operable to communicate with a base station 206 and / or similar wireless infrastructure, which provides access to a computing center 202, e.g., cloud data center. Note that similar results can be obtained with a single computer instead of using a data center 202 or similar computing infrastructure. For example, a laptop can download the collected data via Wifi or from a removable flash memory device. Modem laptop computers can perform similar operations as a data center (e.g., storage, conversion, data reduction / smoothing, analysis, etc.), albeit at a smaller scale.

[0033] The computing center 202 is accessible via the Internet or similar network. A processing block 208 receives data from one or more measurement apparatuses 106. If more than one apparatus 106 provides data from different machines 104, the processingblock 208 can segregate the data so that radio characterization data from different apparatuses 106 can be accessed separately. This may involve, for example, associating a unique ID with each measurement apparatus 106 (and / or machine 104) and indexing each unit of data (e.g., row within a database) with the unique IDs when stored in a database 207. Other data may also be stored along with the radio characterization data 207, such as timestamps, logging messages, other sensor measurements, etc.

[0034] At processing block 209, the received radio characterization data is reduced, smoothed, and / or otherwise processed to provide a well-formatted data set that can be used in subsequent analyses. For example, over a predetermined time period of work and / or after a predetermined number of repetitions of the work, the data for a work region can be collectively processed to average, smooth, interpolate, and otherwise perform statistical analysis (e g., removal of outliers) to obtain a set of representative data that is “well- behaved” for purposes of subsequent processing. An example of subsequent processing is autonomous vehicle simulation at block 210.

[0035] The simulation block 210 is a process for transforming radio characterization data into a navigability map 211. In one embodiment, a navigability map 211 is a location-dependent field indicating a performance score of an autonomous machine grounds maintenance machine. The performance scores rank reliability of the autonomous navigation of the autonomous grounds maintenance machine at the locations. The scores may be provided at any granularity, and may be thresholded to provide a go / no- go, binary indication of navigability within a region. The scores may be based on other factors, such an expected locational error based on the best available navigation mode, assuming the targeted autonomous machine can rely on multiple different navigation modes.

[0036] The simulation 210 may include the application of a transfer function, neural network, or a similar model to the radio characterization map. Generally, the input to the simulation may be any combination of locations and headings along a work path, with the output being an indicator of navigability and / or radio characteristics along the work path. The transfer function may be linear or non-linear, may use spatial filters, convolutions, interpolations, and the like, and utilize other data, such as other maps 214and vehicle models 216. Other types of simulations may be more detailed than a transfer function. For example, a path planning module 218 can use maps 214 that include work region containment boundaries and keep out boundaries. These boundary maps may be user-created (e.g., manual traversal, input to a graphical computer map) and / or may use public domain map boundaries, artificial intelligence (Al) generated boundaries, etc.

[0037] Once a boundary has been established, the path planning module 218 can use machine characteristics from models 216 (e.g., machine speed, machine dimensions, cut / work width, minimum turning radius, etc.) to plan an autonomous path through the work region. Once this is known, a static or dynamic model of the autonomous ground maintenance machine can be simulated as traversing the path. If the generated path crosses a navigation no-go region, then the simulation can used an altered path or change other operational parameters to deal with the loss of wireless-signal-based geolocation in the region along parts of the path. These changes to the work plan (which will be described in detail elsewhere) can be used to iteratively run the simulation until a solution is reached or an error is thrown, e.g., indicating that a full work path cannot be navigated autonomously even if parts of the work region can be navigated autonomously.

[0038] In addition to the indicated performance scores, the navigability map 211 may also include one or more valid work paths that were found by way of the simulation. This work path data may be used for subsequent analysis and tracking, as indicated by follow-up data 212. The follow-up data 212 may include recommendations for improving performance, e.g., optimizing path geometry, indicating a location where a wireless repeater (e.g., RTK signal transmitter) might be added, and the impact of the repeater on navigability (e.g., improvement of X% by area of autonomous coverage). The follow-up data 212 could indicate parts of the work region that can be navigated through but not worked, although could be worked manually using the autonomous ground maintenance machine (e.g., manual assistance) or a using different ground maintenance machine, e.g., an operator controlled work machine. The follow-up data 212 can also include long term updates to the radio signal characterization data 207 that is gathered by the targeted autonomous machine after deployment, so conditions that might affect autonomousperformance (e ., degradation or improvement of a radio base station transmitter) can be monitored to predict whether future changes to the work plan may be needed.[00391 InFIG. 3, a diagram shows a graphical representation of a radio signal characterization map 300 according to an example embodiment. The map 300 includes a location-mapped field of a signal characteristic (in this case RSSI) within the work region 102. Many such maps can be used, such as maps of RSSI for different types of radio signals (satellite GPS, RTK ground station, cellular data), SNR at different bands, etc. A number of radio characteristic maps 300 can be combined with each other and with other maps such as maps of geographical features and / or structures in the work region. The map 300 may be formed by compositing data over time, where the vehicle performs multiple traversals of the work region over a time period to capture varying ambient conditions that affect the radio signal as well as capturing other data that may be useful to autonomous planning, such as operator-chosen work paths and transit routes. The map 300 of the radio signal characteristics in such a case includes a combination of the radio signal characteristics captured during different ones of the multiple traversals. A simulation using this map 300 can form a navigability map 400, and embodiment of which is shown graphically in FIG. 4A.

[0040] This example navigability map 400 provides a binary indication of navigability, e g., navigation is allowed in the shaded areas but not in the unshaded areas. In other embodiments, the navigability map can have more than two indicators (e.g., scale from 1 to 10), and multiple navigability maps can be produced for multiple navigation modes, such as GPS without RTK corrections, RTK-Fixed, RTK -Float, etc. Some modes may be acceptable for traversal of a work sub-region but not for work, in that the machine can navigate reliably enough to get from point A to B (e g., accurate within 20cm or so), but navigation is not deemed to be reliable enough to run a work implement while traversing the sub-region.

[0041] Generally, a map as shown in FIG. 4A indicates a part 402 of work region 102 where the navigability falls below a threshold. Thus when preparing an autonomous work plan to cover the region 102, the navigation through part 402 can be taken into account. In one embodiment, the work region can be split into two work regions 404 thatcan be worked autonomously using the radio navigation source of interest, and another region 402 for which additional measures may be taken.[00421 Inone example, the part 402 where the navigability falls below the threshold can worked manually using the autonomous ground maintenance machine (e.g., assuming the machine can be ridden on, guided via a handle or remote controller) or a different ground maintenance machine (e.g., a manually operated riding mower). This may involve selecting a path for both manual and autonomous work of the respective regions 402, 404 that result in a work pattern (e.g., mowing path) that provides an aesthetically pleasing result. For example, the operator may use a riding mower to mow the part 402 using border-to-border vertical stripes (vertical as seen in the figure) that cover both parts 402, 404, and the autonomous machine can use a path that also uses border-to-border vertical stripes that stay fully within parts 404. The autonomous path may or may not overlap the manually worked regions, e.g., depending on how well the manual paths can be predicted or ascertained and whether such overlap has any deleterious effect on the outcome.

[0043] In other examples, the machine can autonomously work the first parts 404 of the work region in a first navigation mode of the autonomous ground maintenance machine that relies on the radio signal for which the map 400 was created. In these embodiments, the navigability is at or above a threshold in the first parts 404 of the work region. In such a case, the machine autonomously works the second part 402 of the work region in a second navigation mode of the autonomous ground maintenance machine. The second navigation mode does not rely on the radio signal for which the map 400 was created.

[0044] In one example, the second navigation mode for working part 402 may include a different radio signal, e.g., a different RTK base station, a different cellular phone network provider, etc. Such a switchover may occur in software based on crossing a physical boundary based on the navigability map 400. In another example, the second navigation mode may utilize a vision-based navigation system that utilizes cameras attached to the autonomous ground maintenance machines. In order to navigate using vision-based navigation, the autonomous ground maintenance machine will have beenpreviously trained in the second part 402. This training may involve manually guiding the machine through the second part 402 in a patterned or random traversal so that the machine can gather images, build an image feature map of various structures, plants, and other features visible in the part 402, and build a three-dimensional point cloud (3DPC) of the features that can help the machine navigate using cameras. Algorithms such as Simultaneous Location and Mapping (SLAM) and Structure from Motion (SfM) can be used to create a 3DPC in this way. While training in the region and processing the imagery may be time consuming, it may be a viable alternative in cases where the second part 402 is a small fraction of the entire work region.

[0045] In some cases, navigation aids may be manually added or identified that a sensor (e.g., proximity sensor, vision sensor,) can use to navigate using a second mode that does not require the aforementioned radio signals. For example, a marker, beacon, radiofrequency ID tag, above-ground or below-ground marker, boundary wire, etc., may be uniquely identifiable and precisely positioned. When the autonomous ground maintenance machine detects a range and bearing of a pre-identified marker, it can get a fix of its current geolocation. As with vision-based navigation, the placement and / or identification of markers may be more time consuming than relying on wireless geolocation such as GNSS, but can ensure a high level of confidence in GNSS dead zones when it is desired to work the entire work region autonomously.

[0046] In FIG. 4B, a navigability map 410 is shown according to another example embodiment. The map 410 defines zones 412-414 within the work region based on the navigability within each of the zones 412-414 being within respective different ranges of the navigability. In this case, the zones 412-414 are defined in the legend by being suitable for RTK in zone 412, suitable for GPS in zone 413, and suitable for vision-based (or any type of non-GNSS-based navigation) in zone 414. Note that this assumes, for example, a preference order of RTK > GPS > vision-based, so that the most preferable navigation mode in each zone is shown as being usable in the zone if its score is above a threshold. In other words, visual and GPS may also work in zone 412, but since RTK is preferred, that is the selected navigation mode for zone 412.

[0047] One way that the map 410 may be used is that the autonomous ground maintenance machine may be simulated (or directed) to perform the work in at least two of the zones with different scores of the navigability. The work is performed in different modes the autonomous ground maintenance machine for each of the at least two zones. For example, if vision-based navigation is not available or desired, then autonomous work may be performed in just zones 412 and 413. Between those zones, different modes used in the zone may include any combination of different navigation speed, different navigation mode (e.g., GPS or RTK), different traversal paths, and different speed of a work implement used to perform the work. These different modes can account for different capabilities of machine when using the available navigation aids in each zone.

[0048] In FIG. 4C, a navigability map 420 is shown according to another example embodiment. This map 420 includes details of the surroundings of the work region 102, such as sidewalks, trees, buildings, etc. Such details may be overlaid from a public maps database, such as satellite or aerial photograph databases. The work region 102 has two parts 422, 424 that are respectively below or above a navigation threshold for a navigation mode that uses a particular radio signal. The map of the navigability 420 indicates one or more locations 428 where an additional navigation aid can be located to improve the navigability of the autonomous grounds maintenance machine using a navigation mode that does not use the radio signal. The additional navigation aid may include at least one of a beacon, a visual reflective marker, a visual shape marker, a radio reflective marker, and an underground marker. While the locations 428 are shown as discrete positions on the map 420, they may alternatively be represented as a region in which any location may be used.

[0049] Also seen in the navigability map 420 is shaded region 426 that indicates one or more locations where an additional transmitter (e.g., RTK repeater or base station) can be located to improve the navigability of the autonomous grounds maintenance machine. In this case, the transmitter may be placed in the region to sufficiently boost the radio signal in part 422 so that it meets or exceeds the navigability threshold. Some separation between from the work region 102 and the shaded region 426 may be desired, for example, to prevent front end overload of receivers by the additional transmitter. Theother details of the map 420 such as streets, trees, buildings etc. may be accessed from a second map of geographical features to refine the one or more locations for the transmitter based on obstacles shown in the second map.

[0050] The map 420 may be used in some embodiments to indicate tuning parameters for a base station to improve the navigability of the autonomous ground maintenance machine. Tuning the existing base station may preclude the need to add another transmitter. For example, a different transmission spectrum, power output, antenna height or orientation, antenna location, or the like could be added to the simulation in order to predict a change in the size and shapes of the differently navigable parts 422, 424 of the work region 102.

[0051] In some cases, the autonomous ground maintenance machine may be programmed to move through regions for which navigability is below a first threshold for work, but may be above a second threshold for traversal only, e.g., with a work implement turned off. In FIG. 5 A, a diagram shows a machine 104 moving through a first part 502 (shaded part) of a work region 102 in which navigability is above a threshold such that autonomous work is allowed. The lines 501 in the figure indicate a path geometry that the machine 104 is intended to follow through the work region 102. The navigability through part 504 is below a threshold such that it may be inadvisable at least to run a work implement while in this part 504.

[0052] As indicated by paths 506, the machine 104 may be able to traverse the part 504 using dead reckoning, e.g., using wheel encoders to estimate distance and bearing from a known start point. Dead reckoning may be accurate enough such that part 504 of work region 102 can be traversed, e.g., with the work implement turned on or off. There may be a threshold path deviation that allows the work implement to operation. Such threshold may be based on factors such as desired path consistency (e.g., for neatness) and obstacle avoidance (e.g., to avoid damage). For example, if a 5% deviation is estimated for navigation through part 504 and the maximum allowable deviation for work is 10%, then the work implement can be left on when traversing part 504. Other characteristics of the work may also be changed based on the navigation conditions in the part 504, such as vehicle speed, speed of work implement, obstacle avoidance thresholds, etc.

[0053] Once the machine has navigated out of part 504 and wireless geolocation is re-obtained, the work can be continued as indicated by the solid lines to the right of part 504. Note that if expected deviation through part 504 means the work implement is left off or some other criterion is violated within part 504, then paths 508 around part 504 may be used, such that the machine 104 does not leave the shaded part 502. The work implement need not be turned off when using these alternate paths 508 since navigability is expected to be acceptable. Nonetheless, the work implement may still be shut off along parts of paths 508 to prevent overworking overlapping parts of the paths 508.

[0054] Note that because the part 504 only partially segments the work region 102, the paths 506 can be integrated with (e.g., connected to) the work paths, e.g., the solid lines in part 502. In other words, the machine 104 can move along multiple continuous paths between parts 502 and 504. In other embodiments, the autonomous ground maintenance machine 104 can take an abbreviated path through a region with poor navigability, as shown in the block diagram of FIG. 5B. The work machine 104 moves through a first part 512 of a work region 102 in which navigability is above a threshold such that autonomous work is allowed, as indicated by path lines 511. In a second part 514 of the work region 102, navigability is below a threshold such that it may be inadvisable to run a work implement while moving within this part 514.

[0055] Instead of spending a significant amount of time in part 514, the entire first part 512 is worked up to the borders with the second part 514. Then, an abbreviated path 516 is taken across the second part 514, after which a third part 513 is entered where the navigability is again above the threshold. The third part 513 is then worked as indicated by path lines 515. This scenario may be used where the second part 514 has a significant area compared to the other parts 512, 513 such that it is more efficient to avoid multiple traversals of the second part 514, such as paths 506 shown in FIG. 5A. The abbreviated path 516 may be navigated using an alternate means, such as dead reckoning or visual navigation.

[0056] In summary, a navigability map can be formatted in a number of ways to assist in autonomous work planning. For example, a map could indicate which of two or more radios signals are best used for navigating in different parts of a work region. Forexample, if the radio signal includes RTK radio signal, this could involve selecting from two different providers of different signals. The navigability map could be used to indicate one or more locations where an additional RTK transmitter (e.g., wireless repeater) can be located to improve the navigability of the autonomous grounds maintenance machine. In some embodiments, a second map of geographical features of the work region to can be used to refine the one or more locations where the additional RTK transmitter can be located based on obstacles or other features (e.g., convenient structures for mounting a transmitter) shown in the second map. The navigability map may indicate tuning parameters for an RTK transmitter to improve the navigability of the autonomous ground maintenance machine, e.g., a higher location to mount a repeater, use of different spectra from a different provider, etc.

[0057] As described above, a simulation can utilize the navigability map to determine feasible work paths and highlight regions that can be remediated as described above if navigability goes below a threshold. Data from other maps, such as terrain maps, contour maps of natural and manmade structures, and the like, can also be fed into these simulations. Machine gathered data, such as vehicle tilt / pitch measurement, can also be used to build more detailed maps for a similar purpose. This can also be useful for other aspects of autonomous planning, e.g., maximum slope angles of the autonomous ground maintenance machine, effects of terrain on power consumption, etc. The simulations can take into account these other factors to increase realism of the simulation, improve path definition, as well as consider the interaction of other map data on wireless navigation signals.

[0058] In some cases, the path geometry measured during the data gathering may itself be useful in autonomous planning. For example, techniques used by the human operators (e.g., mowing patterns, type of turnaround in headlands) can be used to inform the preparation of autonomous paths. Also, the length of the paths and other factors such as elevation changes in the work region can be used to estimate power consumption by an autonomous vehicle operating along a same or similar work path. Ground slope data can be gathered using on-board vehicle sensors (e.g., inclinometer) via a vehicle telemetry interface as described below, and / or be built into the measuring apparatus.

[0059] In FIG. 6, a diagram illustrates an example of other considerations of a simulation according to an example embodiment. In this example, the simulation includes first and second autonomous machines 104a, 104b. In the illustrated embodiment, both autonomous machines 104a, 104b are ground maintenance machines of a similar type, e.g., mowers. In other embodiments, one of the machines 104a, 104b may be used to provide location services to the other machine but need not to any other work, or may do a different type of work (e.g., one machine mows, the other collects clippings).

[0060] As indicated by the work paths 600, 602, the simulation creates paths for both ground maintenance machines 104a, 104b that are operable to perform part of the work in parallel with each other. Note that if one of the machines 104a, 104b does not work (e.g., operates just as a mobile wireless relay), this machine may be parked at a stationary point or move only so much as needed to maintain radio links.

[0061] For the simulation (and in actual operation), the autonomous machine 104a receives a radio signal 606 from a base station 604 (or other transmitter) and is operable to relay 608 information from the radio signal 606 to the other autonomous ground maintenance machine 104b. Note that the relay signal 608 may be a repeat / retransmission of signal 606 or have the same information as radio signal 606, but operate on a different frequency, modulation scheme, etc. In some embodiments, the relay signal 608 may include the location of the relaying machine 104a, and also include data (or characteristics of the relay signal 608 itself) that allows determining a relative distance and bearing between the two machines 104a, 104b.

[0062] The simulation in the illustrated example would involve path planning for both machines 104a, 104b, taking into account, for example, an maximum allowable separation between the machines 104a, 104b to maintain link via signal 608, locations within the work region 102 where machine 104a can still maintain link via signal 606, intervening terrain, structures, etc. that could block signal 608, etc. Also note that the paths 600, 602 may be different than what is shown. For example, machine 104a may mow using vertical stripes and machine 104b may mow using horizontal stripes. In such a case, the path planner might also consider paths that avoid collisions between machines 104a, 104b.

[0063] In FIG. 7, a block diagram illustrates a system 700 according to an example embodiment. The system 700 includes a vehicle 104 that includes (e.g., is temporarily or permanently equipped with) a measurement apparatus 106. The measurement apparatus 106 includes at least one location sensors 702 and at least one radio receiver 704 that are coupled to a device controller 706. The device controller 706 represents general-purpose or special purpose computing hardware usable to govern the functions of the apparatus 106. The device controller 706 generally includes one or more processors (e.g., central processing units, sub-processor), memory (e.g., random access memory, non-volatile memory), and input / output (I / O) circuits 707 (e.g., I / O busses, general purpose I / O ports, etc.). The device controller 706 may include other devices such as power conditioning circuits, board-monitoring sensors, etc.

[0064] The device controller 706 is configured, via instructions executed by a processor, to repeatedly determine geolocation data via the location sensor 702 and determine radio signal characteristics data corresponding to the respective geolocation data via the radio receiver 704. This gathering of data occurs while the vehicle 104 traverses a work region. These data can be stored in a persistent storage medium 708, such as flash memory and / or magnetic storage. The data storage may be temporary, e.g., buffering or caching before transmittal, or for longer term use, e.g., post-collection analysis, historical records. In some embodiments, no persistent storage may be used on the apparatus 106, e.g., all buffering is done in volatile memory. In such a configuration, the data may be instead persistently stored on a computing arrangement 720, which will be described in greater detail below.

[0065] The measurement apparatus 106 may collect other data in addition to the geolocation data. For example, the apparatus 106 may include a vehicle interface 710 that can interface with the vehicle 104 via a data interface 711 and / or a power interface 713 that communicate with a respective vehicle controller 714 and electrical power system 712. The data interface 711 may use inter-device protocols such as Controller Area Network (CAN), Inter-Integrated Circuit (I2C), etc. The power interface 713 may be a connection to a vehicle electrical power bus, e.g., 12VDC, 24VDC, etc.

[0066] One or both of the data and electrical interfaces 71 1, 713 may allow for collecting data, such as vehicle speed, heading, inclination, activation / deactivation of work units (e.g., blades), inertial measurement unit (IMU) data, power consumption, etc. This vehicle-provided data may be duplicative in some cases of the data measured by the apparatus 106 itself, but can be useful for purposes of cross-checking data for errors, for example. Vehicle-provided data that is different from what is measured by sensors of the apparatus 106 may be used to augment the analyses described elsewhere.

[0067] Data collected by the measurement apparatus 106 can be communicated outside the apparatus 106 via a communications interface 716, which is a software or firmware component that allows transferring data, e.g., to computing arrangement 720. The communications interface 716 may use existing operating system facilities, e.g., remote filesystem application program interface (API), web services client / server interfaces, etc. Alternatively, the communications interface 716 may use a custom protocol, such as streaming data, message passing, etc. Generally, the communications interface 716 communicates via one or more hardware communications interfaces 718. The hardware interface 716 may include wired data transfer such as Universal Serial Bus (USB), wireless data transfer such as WiFi, cellular data, optical communications, etc. The data temporarily or persistently stored on the measurement apparatus 106 may be pushed to or pulled by another device, as represented by computing arrangement 720.

[0068] The computing arrangement 720 may be a large-scale infrastructure computing facility such as an Internet data center, cloud compute facility, etc. The computing arrangement 720 may be one or more single computers, e.g., a laptop, desktop, etc. In some embodiments, the functions described being implemented in the computing arrangement 720 may be performed within the measurement apparatus 106, e.g., by using sufficiently capable processing hardware in the apparatus 106. In such an embodiment, the apparatus 106 may still utilize the communications interface 716 and hardware interface 718 for purposes such as software updates, accessing external map data sources, asset tracking, etc.

[0069] The computing arrangement 720 includes site assessment functions 722 that are intended to capture time-limited data collection activities of one or more measurementapparatuses 106. In this context, “time-limited” refers to the information being collected and used to shape a decision on implementing an autonomous work system. After the decision is made, the collection apparatus 106 may continue to collect data, e.g., to update databases, in a pre-deployment mode. After deployment, an autonomous work vehicle 740 may use similar hardware and software to gather data similar to that provided by the apparatus 106, assuming the vehicles 104, 740 are different.

[0070] The site assessment functions 722 include at least data gathering 724, which involves gradual or bulk receipt of the measurement data from the measurement apparatus 106. The data gathering 724 may also involve data filtering, data smoothing, and other operations to put the data in shape for further use, e.g., by a simulator 726. The processed data is referred to elsewhere as a radio signal characteristics map. The simulator 726 takes data from the measurement apparatus 106 as input along with other data, such as performance parameters of the autonomous vehicle 740, supplementary map data such as terrain maps, etc. The simulator 726 outputs at least one navigability map as described elsewhere herein, which may be stored in a database 723. Note that the radio signal and navigability maps need not strictly conform to the computer science definition of a map, but may be used at least to produce geographically dependent outputs of radio characteristics and navigability scores.

[0071] The simulator 726 may also output work plans 728 based on the simulation, which includes, among other things, traversal paths used by the autonomous machine 740 for performing work. The work plans 728 need not be optimum, but may be at least demonstrate feasibility of autonomous work for a given set of conditions. Once one or more feasible work plans 728 are found, more optimal and / or more detailed operational work plans 732 may be developed for use by an operations function 730, which may operate at the computing arrangement 720, within the autonomous vehicle 740, and / or some other computing system. The operational work plans 732 may consider additional factors, such as minimizing time to completion, energy considerations (e.g., time refueling or recharging), breaking work into segments to fit into work schedules, etc.

[0072] The site assessment functions 722 further include a user interface 725 that provides historical and / or real time data to a user terminal 727. The user terminal 727 mayaccess the user interface 725 via a wide area network, local connection, or may be built into the computing arrangement 720. The user interface 725 provides any combination of graphics, numerical data, and control inputs that allow user-specific functions to be managed via the terminal 727. For example, once a few data gathering session have been completed in a work region, an initial assessment of radio quality may lead to adjustments (e.g., a change in service provider) that can be implemented and followed up in subsequent work session by tracking progress via the user interface 725. This facilitates testing improvements to navigation services before an autonomous system is deployed at the work site.

[0073] The operations function 730 generally manages ongoing work operations with one or more autonomous vehicles 740. This may include providing to the autonomous vehicle 740 a plurality of the work plans 732, such as different paths to apply alternating cutting patterns, different cutting height for different seasons, etc. A site update function may be employed 734 receive data from the autonomous vehicle 740 over time, such that the various maps and work plans can be updated if significant changes are detected. The computing arrangement 720 can communicate with one or both of the measurement apparatus 106 and the autonomous machine 740 via data communications interface hardware 721, e.g., a network adapter. While not shown, the operations function 730 may also have a user interface analogous to user interface 725,

[0074] The autonomous vehicle 740 is shown with data communications interface hardware 741 that is capable of communicating with the computing arrangement 720. The communications with the computing arrangement 720 may be for the purposes of providing full or partial work plans 742, monitoring operations via control function 744, and continuing to provide radio signal measurements at different work locations via ongoing data gathering / updating function 746. The illustrated functions 744, 746 are exemplary, and represent a subset of possible functions that can be used to centrally manage a single autonomous work vehicle or fleet of autonomous work vehicles.

[0075] Note that the work plans 742 may be in the form of dead zone data that provides a geographic description of one of more parts of the work region where the navigability falls below the threshold. For example, one of a number of differentautonomous vehicles may be usable in the work region, and may be able to generate their own custom work paths based, for example, predetermined boundaries of the work region and exclusion zones where the vehicle should avoid during work. The computing arrangement 720 in such a scenario may send the dead zone data (and optionally the work region and exclusion boundaries) to the autonomous vehicle 740. Based on this data, the autonomous vehicle 740 may generate its own path to use during operations. For example, if the autonomous vehicle 740 is configured to work random paths, the dead zone parts of the work region may be treated as exclusion zone, or may be traversed in a way described above, e.g., without performing work, using alternate navigation, using an abbreviated path, etc.

[0076] Note that full or partial work plans 742 may be sent to a target autonomous ground maintenance machine 740 (for which a full plan was created), a different ground maintenance machine, or a dispatch device that communicates to work machines in the area. These other machines may have different performance characteristics than the targeted autonomous grounds maintenance machine 740, and so may develop their own plans using their own local processors and / or send a request to the computing arrangement 720 to develop a new plan customized for the particular characteristics of the different ground maintenance machine or dispatch device.

[0077] In FIG. 8, a flowchart shows a method according to an example embodiment. The method involves traversing 800 a work region with a vehicle. The vehicle is equipped with a radio receiver and a location sensor. While traversing the work region, data are repeatedly determined 801 that include geolocations via the location sensor and radio signal characteristics via the radio receiver. The radio signal characteristics pertain to a radio signal used for navigation assist. A map of the radio signal characteristics to the geolocations is stored 802 in a computer readable medium. Via a computer processor, a simulation may optionally be performed 803 of an autonomous ground maintenance machine traversing the work region to perform work. The simulation includes an effect of the radio signal characteristics on autonomous navigation of the autonomous grounds maintenance machine at the geolocations. A navigability map of the autonomous ground maintenance machine in the work region is determined 804 via the computerprocessor based on the map of the radio signal characteristics and / or the simulation, if the latter is created.[00781 InFIG. 9, a flowchart shows a method according to another example embodiment. The method involves traversing 900 a work region with a vehicle. The vehicle is equipped with a radio receiver and a location sensor. While traversing the work region, data are repeatedly determined 901 that include geolocations via the location sensor and radio signal characteristics via the radio receiver. The radio signal characteristics pertain to a radio signal used for navigation assist. A map of the radio signal characteristics to the geolocations is stored 902 in a computer readable medium. A navigability map of the autonomous ground maintenance machine in the work region is determined 903 via a computer processor. The navigability map is used 904 to plan a work path for the autonomous ground maintenance machine, wherein the work path accounts for regions where the autonomous ground maintenance machine cannot rely on the radio signal for autonomous navigation.

[0079] In view of the above, it will be readily apparent that the functionality of the controllers in the aforementioned apparatuses and systems may be implemented in any manner known to one skilled in the art. For instance, the memory may include any volatile, non-volatile, magnetic, optical, and / or electrical media, such as a random-access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), electrically-erasable programmable ROM (EEPROM), flash memory, and / or any other digital media.

[0080] The processors used in the controllers may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), and / or equivalent discrete or integrated logic circuitry. In some embodiments, the processor may include multiple components, such as any combination of one or more microprocessors, one or more controllers, one or more DSPs, one or more ASICs, and / or one or more FPGAs, as well as other discrete or integrated logic circuitry. The functions attributed to the controller and / or processor herein may be embodied as software, firmware, hardware, or any combination of these. Certain functionality of the controller may also be performed in the cloud or other distributed computing systems operably connected to the processor.

[0081] The inter-device and intra-device communications may use any combination of wired and wireless communications. Examples of wireless data interfaces include WiFi, Bluetooth, cellular modem, inductive data interface, and NFC. Examples of wired interfaces include Universal Serial Bus (USB), Ethernet, Controller Area Network (CAN), Inter-Integrated Circuit (I2C), and serial line (e.g., RS-232, IEEE 1394).

[0082] While the present disclosure is not so limited, an appreciation of various aspects of the disclosure will be gained through a discussion of the specific illustrative aspects provided below. Various modifications of the illustrative aspects, as well as additional aspects of the disclosure, will become apparent herein.

[0083] Example Al is a method comprising traversing a work region with a vehicle, the vehicle equipped with a radio receiver and a location sensor. While traversing the work region, the method comprises repeatedly determining: geolocations via the location sensor; and radio signal characteristics via the radio receiver, the radio signal characteristics corresponding to the respective geolocations and pertaining to a radio signal used for navigation assist. The method further comprises storing a map of the radio signal characteristics at the geolocations in a computer readable medium; and via a computer processor, determining a navigability map of an autonomous ground maintenance machine in the work region based on the map of the radio signal characteristics.

[0084] Example A2 includes the method of example Al, further comprising, via the computer processor, performing a simulation of the autonomous ground maintenance machine traversing the work region to perform work, the simulation including an effect of the radio signal characteristics on autonomous navigation of the autonomous grounds maintenance machine at the geolocations, wherein the navigability map is based on the simulation. Example A3 includes the method of example A2, wherein the simulation further includes simulating a second autonomous machine navigating the work region, the second autonomous machine operable to relay the radio signal to the autonomous ground maintenance machine. Example A4 includes the method of example A3, wherein the second autonomous machine is a second ground maintenance machine operable to perform part of the work in parallel with the autonomous ground maintenance machine.

[0085] Example A5 includes the method of examples A1-A4, wherein traversing the work region with the vehicle comprises manually guiding the vehicle through the work region. Example A6 includes the method of example A5, wherein manually guiding the vehicle comprises performing grounds work in the work region with the vehicle. Example A7 includes the method of any previous example, wherein the radio signal characteristics comprise at least one of signal strength and signal interference. Example A8 includes the method of any previous example, wherein the navigability map includes performance scores as a function of locations in the work region, the performance scores indicating reliability of the autonomous navigation of the autonomous grounds maintenance machine at the locations using the radio signal.

[0086] Example A9 includes the method of any previous example, wherein the radio signal characteristics pertain to two or more radio signals used for the navigation assist. Example A10 includes the method of example A9, wherein the two or more radio signals comprises a real time kinematics (RTK) radio signal and a global navigation satellite system (GNSS) signal. Example Al 1 includes the method of example A9, wherein the navigability map indicates which of the two or more radios signals are best used for navigating in different parts of the work region.

[0087] Example A12 includes the method of any previous example, wherein the radio signal comprises a real time kinematics (RTK) radio signal. Example A13 includes the method of example A12, wherein the navigability map indicates one or more locations where an additional RTK transmitter can be located to improve the navigability of the autonomous grounds maintenance machine. Example A14 includes the method of example A13, further comprising accessing a second map of geographical features of the work region to refine the one or more locations based on obstacles shown in the second map. Example Al 5 includes the method of example A12, wherein the navigability map indicates tuning parameters for an RTK base station to improve the navigability of the autonomous ground maintenance machine.

[0088] Example Al 6 includes the method of any previous example, further comprising determining a work path of the autonomous ground maintenance machine through the work region based on the navigability map. Example Al 7 includes the methodof example Al 6, wherein the work path avoids a part of the work region where the navigability falls below a threshold. Example A18 includes the method of example A17, further causing the autonomous ground maintenance machine to perform the work via the work path, wherein the part of the work region where the navigability falls below the threshold is worked manually using the autonomous ground maintenance machine or a different ground maintenance machine. Example Al 9 includes the method of example Al 6, further causing the autonomous ground maintenance machine to perform the work via the work path, during which the autonomous ground maintenance machine gathers updated radio signal characteristics corresponding to the respective geolocations, the updated radio signal characteristics used to update the map.

[0089] Example A20 includes the method of any previous example, further comprising communicating, to a receiver on an operational machine, a geographic description of one of more parts of the work region where the navigability falls below a threshold. Example A21 includes the method of example A20, wherein the operational machine comprises the autonomous ground maintenance machine, a second ground maintenance machine, or a dispatch device. Example A22 includes the method of example A20, wherein the operational machine comprises a second autonomous ground maintenance machine having different performance characteristics than the autonomous grounds maintenance machine.

[0090] Example A23 includes the method of any previous example, further comprising: autonomously working a first part of the work region in a first navigation mode of the autonomous ground maintenance machine, the navigability being at or above a threshold in the first part of the work region, the first navigation mode relying on the radio signal; and autonomously working a second part of the work region in a second navigation mode of the autonomous ground maintenance machine, the navigability being below the threshold in the second part of the work region, the second navigation mode not relying on the radio signal. Example A24 includes the method of example A23, wherein the second navigation mode relies on a second radio signal different from the radio signal. Example A25 includes the method of example A23, wherein the autonomous ground maintenance machine traverses an abbreviated path through the second part. Example A26 includes themethod of example A25, wherein the second navigation mode comprises dead reckoning. Example A27 includes the method of example A23, wherein the second navigation mode comprises a vision-based navigation. Example A28 includes the method of example A27, wherein the vision-based navigation utilizes an image feature map obtained by training the autonomous ground maintenance machine in the second part.

[0091] Example A29 includes the method of any previous example, wherein the vehicle performs multiple traversals of the work region over a time period to capture varying ambient conditions that affect the radio signal, the map of the radio signal characteristics to the geolocations comprising a combination of the radio signal characteristics captured during different ones of the multiple traversals. Example A30 includes the method of any previous example, wherein the map of the navigability indicates one or more locations where an additional navigation aid can be located to improve the navigability of the autonomous grounds maintenance machine using a navigation mode that does not use the radio signal. Example A31 includes the method of example A30, wherein the additional navigation aid comprises at least one of a beacon, a visual reflective marker, a visual shape marker, a radio reflective marker, and an underground marker.

[0092] Example A32 includes the method of any previous example, further comprising defining zones within the work region based on the navigability within each of the zones being within respective different ranges of the navigability. Example A33 includes the method of example A32, further causing the autonomous ground maintenance machine to perform the work in at least two of the zones with two ranges of the navigability, the work performed in different modes the autonomous ground maintenance machine for each of the at least two zones. Example A34 includes the method of example A33, wherein the different modes comprise any combination of different navigation speed, different navigation mode, different traversal paths, and different speed of a work implement used to perform the work.

[0093] Example A35 is method comprising: traversing a work region with a vehicle, the vehicle equipped with a radio receiver and a location sensor; while traversing the work region, repeatedly determining geolocations via the location sensor anddetermining, via the radio receiver, radio signal characteristics corresponding to the respective geolocations, the radio signal characteristics pertaining to a radio signal used for navigation assist; storing a map of the radio signal characteristics at the geolocations in a computer readable media; via a computer processor, determining a navigability map of an autonomous ground maintenance machine in the work region based on the map of the radio signal characteristics; and using the navigability map to plan a work path for the autonomous ground maintenance machine, wherein the work path accounts for regions where the autonomous ground maintenance machine cannot rely on the radio signal for autonomous navigation.

[0094] Example A36 includes the method of example A35, wherein the work path avoids a part of the work region where the navigability falls below a threshold. Example A37 includes the method of examples A35 or A36, further comprising: working a first part of the work region in a first navigation mode of the autonomous ground maintenance machine, the navigability being at or above a threshold in the first part of the work region, the first navigation mode relying on the radio signal; and working a second part of the work region in a second navigation mode of the autonomous ground maintenance machine, the navigability being below the threshold in the second part of the work region, the second navigation mode not relying on the radio signal.

[0095] Example A38 is a system, comprising: a vehicle comprising a measurement apparatus, the measurement apparatus including a radio receiver and a location sensor coupled to a system controller, wherein the measurement apparatus is operable to, while the vehicle traverses a work region, repeatedly determine geolocation data via the location sensor and determine, via the radio receiver, radio signal characteristics data corresponding to geolocations of the geolocation data, the radio signal characteristics data pertaining to a radio signal used for navigation assist; and a computing arrangement configured to receive the geolocation data and the radio signal characteristics data from the measurement apparatus, the computing arrangement further operable to: store the radio signal characteristics data and the geolocation data in a computer readable medium; perform a simulation of an autonomous ground maintenance machine traversing the work region to perform work, the simulation using the radio signal characteristics data and the geolocationdata simulating an effect of radio signals on autonomous navigation of the autonomous grounds maintenance machine in the work region; and determine a navigability map of the autonomous ground maintenance machine in the work region.

[0096] Example A39 includes the system of example A38, wherein the computing arrangement is further configured to determine a work path of the autonomous ground maintenance machine through the work region based on the navigability map, wherein the work path accounts for regions where the autonomous ground maintenance machine cannot rely on the radio signal for autonomous navigation. Example A40 includes the system of example A39, further comprising the autonomous ground maintenance machine, the autonomous ground maintenance machine operable to: receive work path data describing the work path via a data interface; and autonomously work in the work region via the work path.

[0097] It is noted that the terms “have,” “include,” “comprises,” and variations thereof, do not have a limiting meaning, and are used in their open-ended sense to generally mean “including, but not limited to,” where the terms appear in the accompanying description and claims. Further, “a,” “an,” “the,” “at least one,” and “one or more” are used interchangeably herein. Moreover, relative terms such as ’’left,” “right,” “front,” “fore,” “forward,” “rear,” “aft,” “rearward,” “top,” “bottom,” “side,” “upper,” “lower,” “above,” “below,” “horizontal,” “vertical,” and the like may be used herein and, if so, are from the perspective shown in the particular figure, or while the machine is in an operating configuration. These terms are used only to simplify the description, however, and not to limit the interpretation of any embodiment described. As used herein, the terms “determine” and “estimate" may be used interchangeably depending on the particular context of their use, for example, to determine or estimate a position or pose of a vehicle, boundary, obstacle, etc.

[0098] Further, it is understood that the description of any particular element as being connected to or coupled to another element can be directly connected or coupled, or indirectly coupled / connected via intervening elements.

[0099] Unless otherwise indicated, all numbers expressing feature sizes, amounts, and physical properties used in the specification and claims are to be understood as beingmodified in all instances by the term “about.” Accordingly, unless indicated to the contrary, the numerical parameters set forth in the foregoing specification and attached claims are approximations that can vary depending upon the desired properties sought to be obtained by those skilled in the art utilizing the teachings disclosed herein. The use of numerical ranges by endpoints includes all numbers within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5) and any range within that range.

[0100] The various embodiments described above may be implemented using circuitry, firmware, and / or software modules that interact to provide particular results. One of skill in the arts can readily implement such described functionality, either at a modular level or as a whole, using knowledge generally known in the art. For example, the flowcharts and control diagrams illustrated herein may be used to create computer-readable instructions / code for execution by a processor. Such instructions may be stored on a non- transitory computer-readable medium and transferred to the processor for execution as is known in the art. The structures and procedures shown above are only a representative example of embodiments that can be used to provide the functions described hereinabove.

[0101] Note that any components described herein using terms such as “processor,” “controller,” “logic circuit,” “CPU,” or the like may be implemented using a plurality of discrete units operating together. For example, a processer that performs a series of steps or operations may be construed as two or more processors operating cooperatively to perform the steps. Similarly, other processing hardware such as memory and input-output may perform the described functions with multiple discrete units operating cooperatively or being coordinated by another unit, e.g., by a central processor or processors.

[0102] The foregoing description of the example embodiments has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Many modifications and variations are possible in light of the above teaching. Any or all features of the disclosed embodiments can be applied individually or in any combination and are not meant to be limiting, but purely illustrative. It is intended that the scope of the invention be limited not with this detailed description, but rather determined by the claims appended hereto.

Claims

CLAIMS:

1. A method comprising: traversing a work region with a vehicle, the vehicle equipped with a radio receiver and a location sensor; while traversing the work region, repeatedly determining: geolocations via the location sensor; and radio signal characteristics via the radio receiver, the radio signal characteristics corresponding to the respective geolocations and pertaining to a radio signal used for navigation assist; storing a map of the radio signal characteristics at the geolocations in a computer readable medium; and via a computer processor, determining a navigability map of an autonomous ground maintenance machine in the work region based on the map of the radio signal characteristics.

2. The method of claim 1, further comprising, via the computer processor, performing a simulation of the autonomous ground maintenance machine traversing the work region to perform work, the simulation including an effect of the radio signal characteristics on autonomous navigation of the autonomous grounds maintenance machine at the geolocations, wherein the navigability map is based on the simulation.

3. The method of claim 2, wherein the simulation further includes simulating a second autonomous machine navigating the work region, the second autonomous machine operable to relay the radio signal to the autonomous ground maintenance machine.

4. The method of claim 3, wherein the second autonomous machine is a second ground maintenance machine operable to perform part of the work in parallel with the autonomous ground maintenance machine.

5. The method of any one of claims 1-4, wherein traversing the work region with the vehicle comprises manually guiding the vehicle through the work region.

6. The method of claim 5, wherein manually guiding the vehicle comprises performing grounds work in the work region with the vehicle.

7. The method of any one of claims 1-6, wherein the radio signal characteristics comprise at least one of signal strength and signal interference.

8. The method of any one of claims 1-7, wherein the navigability map includes performance scores as a function of locations in the work region, the performance scores indicating reliability of the autonomous navigation of the autonomous grounds maintenance machine at the locations using the radio signal.

9. The method of any one of claims 1-8, wherein the radio signal characteristics pertain to two or more radio signals used for the navigation assist, wherein the two or more radio signals comprises a real time kinematics (RTK) radio signal and a global navigation satellite system (GNSS) signal.

10. The method of claim 9, wherein the navigability map indicates which of the two or more radios signals are best used for navigating in different parts of the work region.

11. The method of any one of claims 1-10, wherein the radio signal comprises a real time kinematics (RTK) radio signal, wherein the navigability map indicates one or more locations where an additional RTK transmitter can be located to improve the navigability of the autonomous grounds maintenance machine.

12. The method of any one of claims 1-11, further comprising determining a work path of the autonomous ground maintenance machine through the work region based on the navigability map.

13. The method of claim 12, further causing the autonomous ground maintenance machine to perform the work via the work path, during which the autonomous ground maintenance machine gathers updated radio signal characteristics corresponding to the respective geolocations, the updated radio signal characteristics used to update the map.

14. The method of any one of claims 1-13, wherein the map of the navigability indicates one or more locations where an additional navigation aid can be located to improve the navigability of the autonomous grounds maintenance machine using a navigation mode that does not use the radio signal.

15. A system, comprising: a vehicle comprising a measurement apparatus, the measurement apparatus including a radio receiver and a location sensor coupled to a system controller, wherein the measurement apparatus is operable to, while the vehicle traverses a work region, repeatedly determine geolocation data via the location sensor and determine, via the radio receiver, radio signal characteristics data corresponding to geolocations of the geolocation data, the radio signal characteristics data pertaining to a radio signal used for navigation assist; and a computing arrangement configured to receive the geolocation data and the radio signal characteristics data from the measurement apparatus, the computing arrangement further operable to: store the radio signal characteristics data and the geolocation data in a computer readable medium; perform a simulation of an autonomous ground maintenance machine traversing the work region to perform work, the simulation using the radio signal characteristics data and the geolocation data simulating an effect of radio signals on autonomous navigation of the autonomous grounds maintenance machine in the work region; anddetermine a navigability map of the autonomous ground maintenance machine in the work region.

16. The system of claim 15, wherein the computing arrangement is further configured to determine a work path of the autonomous ground maintenance machine through the work region based on the navigability map, wherein the work path accounts for regions where the autonomous ground maintenance machine cannot rely on the radio signal for autonomous navigation.

17. The system of claim 16, further comprising the autonomous ground maintenance machine, the autonomous ground maintenance machine operable to: receive work path data describing the work path via a data interface; and autonomously work in the work region via the work path.

Images