Apple won 58 Patents today covering an Autonomous Vehicle Lane Change System and an Odd one covering future Wearables
Today the U.S. Patent and Trademark Office officially published a series of 58 newly granted patents for Apple Inc. In this particular report we briefly cover a Project Titan patent relating to an autonomous vehicle lane change system along with a unique, yet odd, wearables patent. And as always, we wrap up this week's granted patent report with our traditional listing of the remaining granted patents that were issued to Apple today.
Patent #1
Autonomous Vehicle Lane Change
In Apple's patent background Apple notes that automated vehicle control systems are being developed that take on more and more of the work of maneuvering a vehicle. Automated vehicle control systems are able to access and integrate a wealth of information that is practically unavailable to human operators, such as data from a variety of sensors or communications received from other vehicles or traffic control systems, which fundamentally changes the vehicle control problem and provides an opportunity to automatically maneuver vehicles in a far safer and more efficient manner. However, the real world is highly complex and it is challenging to design automated control systems that can robustly handle the wide variety of scenarios that may be encountered on a road with other automated and/or human operated vehicles.
Apple's patent covers lane change maneuvers that are important for autonomous vehicles. The patent describes an approach for an integrated decision making and trajectory determination solution for lane change maneuver motion planning. The decision making component decides the behavioral pattern to be followed by an autonomous vehicle and the trajectory determination component computes the actual lane change trajectory for performing the maneuver. Note that the decision making component and the trajectory component may work in a symbiotic manner to enable a successful lane change.
The lane change problem may be specified using input parameters including: (1) a current kinematic state (e.g., position, orientation, linear velocities, and angular velocities) of an ego vehicle to be maneuvered; (2) a planning horizon in time (the duration of motion plan); (3) an origin lane (i.e., the lane in which the ego vehicle starts the maneuver) and a target lane (i.e., the lane into which the ego vehicle will transition during the lane change); and (4) predictions for other static/dynamic obstacles in the origin and target lanes for the duration of the planning horizon).
The decision making component either decides that a desired lane change is not possible, i.e., the problem is unsolvable under the current constraints, or chooses a gap in the target lane into which ego vehicle can maneuver. When a lane change is feasible, the decision making component may provide the trajectory determination component with an identification of the chosen gap, a temporal parametrization of the phases of the lane change maneuver, and an initial guess for the acceleration profile to be used during the lane change maneuver.
This approach to lane change problem includes adjusting the lane boundary constraints such that the vehicle uses the ego lane corridor during a preparation phase before a start time, uses both lanes to perform the transition during an execution phase in a time window between the start time and an end time, and finally only uses the target lane during a completion phase for computing the rest of the plan. We also use an attract/repel mechanism for smooth transition. During the three different phases of the lane change maneuver we enforce distance based and physics based separation constraints between the ego vehicle and vehicles in the origin lane and target lane. Given the kinematic state of the ego vehicle and predictions for other static/dynamic objects in the origin lane and target lane, the decision making component identifies a gap into which to lane change, a temporal parametrization of the lane change maneuver, and an initial guess for the acceleration profile. This is used by trajectory determination component to compute the final motion plan.
Using this approach for motion planning lane changes may offer a number of advantages over other approaches that use a uniform planner irrespective of the planning maneuver. When a uniform planner is used, the lane change maneuver differs from others (e.g., in lane driving, planning for intersection), only in the manner that the problem is setup (e.g., specific targets and behavioral restrictions). In contrast, this approach uses a new planner for lane change which uses different cost functions, maneuver windows and constraints suited for lane change. This approach may help in a variety of ways. First, as this planner is specialized for lane change, we need not adjust its behavior to cater to other planning problems (e.g., in lane motion), which means each planner can optimize its behavior independently and not requiring to adjust for a uniform solution. Second, this approach provides the flexibility to evaluate different maneuver solutions (computed in parallel) and choose the best suited one at a higher level. Third, the specialized optimization-based approach allows for broader set of operational conditions as it increases flexibility by identifying a lane change maneuver in terms of a homotopy class and parametrization, rather than only attempting to compute a lane change maneuver using a fixed homotopy (e.g., the closest gap to the ego vehicle) or fixed parametrization (e.g., only considering maneuvers that are executable with a fixed start time rather than making the start time a parameter of the problem that is determined dynamically).
Apple's patent FIG. 1 below is an illustration of an example of a lane change maneuver of a vehicle; FIG. 3 is an illustration of an example of lane change scenario: FIG. 4 is a block diagram of an example of a vehicle configured for lane change control.
For more details, check out Apple's granted patent #US 11580859 B1.
Patent #2
Interesting, yet Odd Wearables Patent
Apple was granted an oddball patent today that covers a possible next-gen Apple Watch or new wrist band device that could capture a user's wrist, arm, ankle tattoo and when the watch covers the tattoo, the display will not only present the tattoo on the face of the watch but also enhance it with added color or in animation mode as an option.
Apple's patent FIGS. 7A, B, C, D, G and J below illustrate example user interfaces for updating a display of a device relative to a user's body; After the watch captures the tattoo, it could reset it so it resides on the face of the watch in its natural top wrist position as captured in Apple's patent FIG. 7E below.
Apple notes that as the user removes their watch or wrist device, the tattoo goes into blur mode of the tattoo and then shuts off as noted in FIG. 7J above.
While that part of the invention could be rather interesting for those with tattoos, a second phase of the patent is a real oddity. Some will find it interesting if not practical while others will think of it as being a bit creepy.
A Bit Odd: While Apple didn't provide patent figures for this next phase, it goes like this: The principle of the patent isn't restricted to tattoos for a watch. For instance, if a user puts on Apple's future mixed reality headset, the internal camera(s) could capture the user's eyes, nose, eyebrows, hair, glasses, freckles and other facial features and present them on the face of the headset device.
It does nothing for the user of course, but for those in their environment, they'll be able to distinguish who has the headset on. Once you can visualize that scenario, it become rather ingenious. This would allow a user to walk on the street as well with their headset and the public would be able to see the rest of your face behind your headset face plate.
Apple's patent states: "Updating a display to show a portion of a user's body that is hidden behind the electronic device provides a more intuitive way for the user, and others in the surrounding environment who view the display, to know where the electronic device is located relative to the user's body, even when that portion of the user's body is covered and not visible to the user (or others in the surrounding environment). This improves the experience of interacting with wearable devices by decreasing the cognitive burden on the user, and others in the surrounding environment, to imagine or guess what is behind the electronic device."
If this interests you, check out granted patent US 11579693 B2
Today’s Remaining Granted Patents
Comments