The First Apple Patent Regarding Autonomous Vehicle Navigation was published today by the U.S. Patent Office
Today the US Patent & Trademark Office published a patent application from Apple that relates to autonomous navigation of a vehicle, and in particular to development and evaluation of an autonomous navigation route characterization which can be utilized by at least some portion of a vehicle to autonomously navigate the route.
This is Apple's first patent on Autonomous vehicle technology and for that Patently Apple has opened a new archive category as this is likely the first of many patents to come on this subject.
Apple begins with an overview that notes that the rise of interest autonomous navigation of vehicles, including automobiles, has resulted in a desire to develop autonomous navigation systems which can autonomously navigate (i.e., autonomously "drive") a vehicle through various routes, including one or more roads in a road network, such as contemporary roads, streets, highways, etc. However, systems which can enable autonomous navigation, also referred to as autonomous driving, of a vehicle can be less than ideal.
In some cases, autonomous navigation is enabled via an autonomous navigation system which can process and respond to static features (e.g., roadway lanes, road signs, etc.) and dynamic features (present locations of other vehicles in a roadway on which the route extends, present environmental conditions, roadway obstructions, etc.) along a route in real-time as they are encountered, thereby replicating the real-time processing and driving capabilities of a human being. However, the processing and control capabilities required to simulate such processing and responsive capability can be impractical, if technically feasible, and the complexity and magnitude of computer systems required to be included in a vehicle to enable such real-time processing and responsiveness can present an unsuitably excessive investment in capital costs for each vehicle, thereby rendering the system impractical for usage on a wide scale.
In some cases, autonomous navigation is enabled by developing a detailed map of various routes, including data indicating various features of the road (e.g., road signs, intersections, etc.), specifying various driving rules relative to the various routes (e.g., proper speed limits, lane changing speeds, lane locations, variations of driving rules based on various climate conditions and times of day, etc. for a given portion of a given route), and providing the map to autonomous navigation systems of various vehicles to enable the vehicles to autonomously navigate the various routes using the map.
However, development of such a map can require extensive expenditures of time and effort, as developing sufficient data for an individual route can require dispatching a suite of sensors, which can be mounted in a dedicated sensor vehicle, to traverse a route and collect data regarding the various features included in the route, processing the collected data to develop a "map" of the route, determining appropriate driving rules for various portions of the route, and repeating the process for each of the individual routes included in the map. Such a process can require an excessive expenditure of time and effort to develop a map characterizing multiple routes, particularly when the multiple routes span over some or all of the roadways in a major city, region, nation, etc.
In addition, as roadways can change over time, e.g. due to road construction, accidents, weather, seasonal occurrences, etc. such a map can unexpectedly become obsolete and unusable for safe autonomous navigation of a route. Updating a map can require dispatching a sensor suite to re-traverse the route, which can require an expenditure of time. When such an expenditure is considered in view of the sheer volume of potential routes in a roadway network, particularly if multiple routes require updating simultaneously, updating a map of routes in a timely manner, such that vehicle users are not deprived of safe autonomous navigation capabilities, can be difficult.
Apple's invention provides a vehicle configured to autonomously navigate a driving route. The vehicle includes sensor devices which monitor characteristics of the driving route based on the vehicle being navigated along the driving route, and an autonomous navigation system which is interoperable with the sensor devices to:
"implement a succession of updates to a virtual characterization of the driving route, based on monitoring a succession of manual navigations of the vehicle along the driving route, associate a confidence indicator with the virtual characterization, based on monitoring the succession of updates to the virtual characterization, and enable autonomous navigation of the vehicle along the driving route, based at least in part upon a determination that the confidence indicator at least meets a threshold confidence indication, such that the autonomous navigation system autonomously navigates the vehicle along at least a portion of the driving route, based on controlling one or more control elements of the vehicle and based on a user-initiated command, received at the autonomous navigation system via a user interface of the vehicle, to engage in autonomous navigation of the portion of the driving route."
Some embodiments provide an apparatus which includes an autonomous navigation system configured to be installed in a vehicle and selectively enable autonomous navigation of the vehicle along a driving route. The autonomous navigation system can include a route characterization module which implements a succession of updates to a virtual characterization of the driving route, wherein each update is based on monitoring a separate one of a succession of manually-controlled navigations of the vehicle along the driving route, and implementing each update of the succession of updates includes associating a confidence indicator with the virtual characterization based on a monitored variation, of the virtual characterization which is associated with the respective update. The autonomous navigation system can include a route evaluation module configured to enable user-initiated autonomous navigation of the driving route by the vehicle, based on a determination that a confidence indicator associated with the characterization of the driving route exceeds a threshold confidence indication.
Other embodiments provide a method which includes performing, by one or more computer systems installed in a vehicle: receiving a set of sensor data associated with a driving route, from a set of sensors included in the vehicle, based at least in part upon the vehicle being manually navigated along the driving route, processing the set of sensor data to update a stored characterization of the driving route, wherein the stored characterization is based on at least one previously-generated set of sensor data associated with one or more historical manual navigations of the vehicle along the driving route associating a confidence indicator with the updated characterization based on a comparison of the updated characterization with the stored characterization, and enabling availability of user-initiated autonomous navigation, by the vehicle, of the driving route, based at least in part upon a determination that the confidence indicator at least meets a predetermined threshold confidence indication.
Apple's patent FIG. 1 illustrated below shows a schematic block diagram of a vehicle #100 which comprises an autonomous navigation system (ANS); FIG. 2 illustrates an illustration of a vehicle, which includes an ANS and a set of sensor devices, navigating through a region which includes multiple roadway portions of multiple roadways
Apple's patent FIG. 4 presented below illustrates a block diagram of an autonomous navigation system (ANS); FIG. 5A illustrates a user interface associated with the autonomous navigation system; FIG. 12 illustrates implementing at least a portion of a curation spectrum with regard to one or more virtual characterizations, of one or more roadway portions.
Apple's patent FIG. 6 below illustrates another user interface associated with the autonomous navigation system; FIG. 9A illustrates a schematic of an autonomous navigation network.
Route Characterization Development
The following is covering patent FIG. 2 in more detail.
In some embodiments, an Autonomous Navigation System (ANS) can develop one or more virtual characterizations of one or more roadway portions, which the ANS can subsequently utilize to autonomously navigate a vehicle through the one or more roadway portions, based on monitoring various static features, dynamic features, driving characteristics, etc. while the vehicle is navigated through the one or more roadway portions. Such monitoring can be implemented when the vehicle is manually navigated through the one or more roadway portions by a vehicle user, such that the ANS can develop a characterization of the static features of a route by monitoring the static features when the vehicle is manually navigated along the route and can develop a set of driving rules specifying how ANS is to navigate a vehicle through the one or more roadway portions based on monitoring driving characteristics of the user's manual navigation of the vehicle through the roadway portion, monitoring driving characteristics of other vehicles navigating through the roadway portion in proximity to the local vehicle, etc.
As a result, a vehicle ANS can both develop a characterization of the physical state of the route (e.g. static features) and a characterization of how to navigate along the route (e.g., driving rules) based on monitoring a manual navigation along the route and various features observed while the vehicle is manually navigated through the route. As a result, the ANS can develop characterizations used to engage in autonomous navigation of a route, independently of externally-received or preexisting characterization data.
Apple's patent FIG. 2 illustrates an illustration of a vehicle #202, which includes an ANS #201 and a set of sensor devices #203, navigating through a region #200 which includes multiple roadway portions #210A-D of roadways #208, #218, according to some embodiments.
The vehicle can be manually navigated through the route, and sensor devices can include one or more external sensor devices, vehicle sensor devices, etc. The vehicle and ANS can be included in any of the embodiments of a vehicle, ANS, etc.
As shown in the illustrated embodiment, a region #200 which includes one or more various roadways #208, #218 can be divided into various roadway "portions" #210. An ANS can demarcate various roadway portions based on position data received from one or more position sensors in the vehicle, one or more various static features included in the region etc.
Different roadway portions can have different sizes, which can be based at least in part upon driving velocity of vehicles navigating the roadway in which the roadway portions are included, environmental conditions, etc. For example, roadway #208 can be a highway where the average driving velocity is higher than that of roadway #218 which can be an onramp; as a result, each of roadway portions #210A-C of roadway #208 can be larger than roadway portion #210D of roadway 218.
In some embodiments, as a vehicle is navigated (manually, autonomously, etc.) through one or more various roadways, an ANS included in the vehicle monitors various static feature characteristics of various roadway portions of the various roadways, monitors various driving characteristics of the vehicle user, other proximate vehicles, etc., as the vehicle navigates through the various roadway portions, some combination thereof, and the like. Such monitoring, which can be implemented by ANS based on input data received from the sensors, can include processing the various characteristics to develop virtual characterizations of the one or more roadway portions through which the vehicle is navigated. The ANS can subsequently utilize the virtual characterizations to engage in autonomous navigation of the vehicle through the one or more roadway portions.
In some embodiments, monitoring of various static feature characteristics of a roadway portion includes identifying various static features associated with the roadway portion. For example, in the illustrated embodiment, where vehicle #202 is navigating through roadway portion #210B of roadway #208, the sensor devices #203 can monitor various aspects of the external environment of region #200 to identify various static features associated with the roadway portion #210B, including the edges #212A-B, lane boundaries #217A-B, lanes #214A-C of roadway #208.
In some embodiments, one or more sensor devices can identify the material composition of one or more portions of roadway #208. For example, a sensor device of the vehicle can include an internal sensor device which can monitor dynamics of the turning of the wheels of the vehicle 202 to determine whether the vehicle is presently navigating over an asphalt surface, gravel surface, concrete surface, dirt surface, etc.
In some embodiments, the sensor devices can monitor various aspects of the external environment of region #200 to identify various static features associated with the roadway portion #210B of roadway which are external to the roadway #208 itself, including static landmarks #213, natural environmental elements #215, road inclines #242, road signs #221, #223, etc.
In some embodiments, identifying a static feature includes identifying information associated with the static feature, including identifying information presented on a road sign. For example, region #200 includes road signs #221, #223, where road sign #221 indicates the presence of onramp #218 and road sign #223 is a speed limit sign which indicates a speed limit for at least roadway portion #210B.
Monitoring static features associated with roadway portion #210B as the vehicle navigates through the portion #210B includes the ANS, based on monitoring the external environment in region #200, determining the physical location of road signs #221, #223 in the roadway portion #210B, identifying the information presented on the road signs #221, #223, and including such information as part of the virtual characterization of the roadway portion.
For example, ANS #201 can, based on monitoring of region #200 by the sensors as the vehicle navigates through roadway portion #210B, identify the physical position of road sign #223 in portion #210B, identify that road sign #223 is a speed limit sign, identify the speed limit indicated by the road sign as 55 miles/hour, and incorporate such information into a driving rule characterization associated with at least the roadway portion #210B as a maximum driving velocity when navigating through at least portion #210B.
In some embodiments, sensor devices 203 can monitor driving characteristics of vehicle #201 and other vehicles #232-236 navigating through roadway portions #210 proximate to vehicle #202, etc. Such driving characteristics can be utilized by the ANS to develop one or more portions of the virtual characterization of one or more roadway portions, including one or more static feature characterizations, driving rules, etc.
For example, based on monitoring driving velocity of one or more of the vehicles navigating through roadway portions #210A-C, the ANS can determine a driving velocity range for autonomously navigating through the one or more roadway portions. The ANS can determine, based on monitoring driving characteristics of the one or more vehicles, a permissible range of acceleration rates associated with navigating through particular roadway portions, locations in the roadway portions where acceleration events are likely, a location of lanes #214A-C in the roadway, a permissible range of spacing distances #252, #254, between vehicle #202 and other vehicles navigating one or more roadway portions #210A-C in a common lane #214 as vehicle #202, a permissible range of spacing distances #256A-B between vehicle #202 and one or more boundaries of the lane #214B in which the vehicle #202 is navigating, etc.
Apple's patent application was filed back in Q2 2017. Considering that this is a patent application, the timing of such a product to market is unknown at this time.
Patently Apple presents a detailed summary of patent applications with associated graphics for journalistic news purposes as each such patent application is revealed by the U.S. Patent & Trade Office. Readers are cautioned that the full text of any patent application should be read in its entirety for full and accurate details. About Making Comments on our Site: Patently Apple reserves the right to post, dismiss or edit any comments. Those using abusive language or negative behavior will result in being blacklisted on Disqus.