Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.
1.2Waypoints from surveyed centerlines
What ifWhat if waypoint placement uses a surveyed lane centerline and the vehicle's map position?
Claim language“based on lane boundary information”
A map stores centerline geometry recorded by a survey vehicle. The controller uses the existing route's lane identifiers to select the centerline segments and samples the portion ahead of the vehicle at one-meter intervals.
1.4Clearance from roadside obstacles
What ifWhat if the lateral adjustment moves route points toward the position with the greatest clearance from detected obstacles?
Claim language“toward a lane center”
An occupancy grid marks parked vehicles and other obstacles, and lane boundaries limit the candidate lateral positions. At each route station, the controller selects the position with the greatest clearance from occupied cells and computes the shift from the current point.
1.5Joint optimization of route points
What ifWhat if the planner solves for the final point positions in one constrained optimization?
Claim language“based on the first adjustment value and the second adjustment value”
The solver treats the lateral coordinates of all route points as decision variables. It minimizes a single objective that penalizes curvature variation and lane-center error while enforcing lane boundary limits.
1.6Route encoded as polynomials
What ifWhat if the updated route stores polynomial coefficients as its geometry?
Claim language“comprising the adjusted plurality of route points”
The planner fits piecewise quintic polynomials to the adjusted points, matching position and the first two derivatives at segment joins. It stores the coefficients and each segment's parameter interval so the controller can evaluate the position, heading, and curvature at a requested station.
1.7Direct steering command output
What ifWhat if the apparatus outputs a steering angle request derived from the route it holds internally?
Claim language“a signal indicating the updated driving route”
The processor reads the route from local memory and computes a steering angle from the vehicle pose and a lookahead point. It sends the requested angle to an actuator controller, which uses measured steering angle to close the actuator loop.
Section 2 · machine-generated
Variations
Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.
Other field of use
Route smoothing for warehouse carts
An overhead camera maps painted aisle boundaries into floor coordinates and sends them to a cart controller that stores a route through the aisle. The controller inserts route points where the aisle boundaries change direction and computes two lateral offsets at each point. One offset reduces differences in curvature between adjacent points, and the other pulls the point toward the mapped aisle centerline. The controller combines the offsets, limits each shift to maintain clearance for the cart body, and sends the resulting route to a steering controller.
Component substitution
Cubic spline for curvature correction
A cubic smoothing spline supplies the first adjustment value from route points selected using lane boundary data. The processor fits the spline to the points in local road coordinates, with a penalty on the integral of the squared second derivative of lateral position along the route. It uses the lateral residual between each original point and the spline as the first adjustment and the offset to a stored lane centerline as the second. The processor blends the offsets, clips each shift to preserve boundary clearance, and outputs the updated points to the steering controller.
Parameter or scale variation
Speed dependent route horizon and spacing
A processor runs the route update at 20 Hz and examines the next 2.5 seconds of travel, with the resulting distance limited to 12 to 60 meters. It places points every 1 meter on segments with absolute curvature above 0.02 inverse meters and every 3 meters elsewhere. It estimates curvature changes using the actual distances between points, blends the resulting smoothing offsets with lane centering offsets, and limits the point shifts to the lane corridor. A tracking controller receives the updated route at each cycle and computes steering commands.
Short horizon replanning for docking lanes
A controller for a vehicle traveling below 2 meters per second updates the route within an 8 meter docking corridor at 10 Hz. It inserts points at 0.25 meter intervals over the final 2 meters and at 1 meter intervals along the approach, with the terminal position and tangent fixed to the docking location and heading. It combines curvature smoothing and lane centering offsets at the interior points, limits each shift to 0.05 meter per cycle, and checks the vehicle footprint against both boundaries. A low speed tracking controller follows the corrected route and stops at the terminal point.
Combination with existing systems
Route updates with predictive speed control
A lateral route planner combines curvature smoothing and lane centering offsets, then sends corrected points and their estimated curvature to a model predictive speed controller. The speed controller uses the corrected route to predict lateral acceleration from speed squared times absolute curvature. It computes a speed profile subject to lateral acceleration, braking, and acceleration limits. A steering controller tracks the corrected points while the speed controller sends commands to the propulsion and brake actuators.
Automated or data-driven variant
Sensor confidence controls lane centering
A forward camera estimates painted boundary positions and their uncertainty, while a scanning range sensor measures raised road edges. The processor fuses the measurements in vehicle coordinates, fits a lane corridor, and propagates boundary uncertainty to the estimated centerline. At each route point it computes a curvature smoothing offset and a centerline offset, then reduces the weight of the centerline offset as estimated uncertainty increases. It constrains the updated points to the usable corridor and sends the route to the steering controller.
Specific version of a part
Direct fitting of clothoids
A curve solver joins the start and end poses with clothoid segments and matches curvature at each join. The solver samples the resulting curve by arc length to produce route points.
Multicore automotive microcontroller
A multicore microcontroller runs lane geometry and route planning tasks on separate cores. A hardware timer releases the planning task every 20 milliseconds, and on-chip memory holds the working route arrays.
Boundary indexed route sampling
The processor pairs left and right lane boundary observations by longitudinal station and identifies the span of the existing route covered by both boundaries. It places route points along that span at one-meter intervals and adds a point at the end of the span.
Weighted lateral point updates
At each route point, the processor expresses the curvature-smoothing adjustment and the lane-centering adjustment as signed displacements along the local route normal. It adds 70 percent of the smoothing displacement and 30 percent of the centering displacement, then clamps the result to the available lane clearance.
Curvature feedforward steering control
The steering controller reads the updated route from the signal and computes a feedforward steering angle from the wheelbase and the route curvature. It adds feedback terms for lateral position error and heading error, then applies a steering rate limit before sending the command to the actuator.
Section 3 · quoted from USPTO records
Other work
Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.
1.2a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: set, based on lane boundary information of a driving lane of the vehicle, a plurality of route points on an existing driving route of the vehicle,
One constraint of the optimization problem can be that the longitudinal displacement of the points of the path sought must be equal to the longitudinal displacement of the points in the reference path, in order to ensure the smoothing of the refined path.
B · US 2024/0034311 A1 · ¶ 19
1.3a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: determine a first adjustment value for reducing amounts of curvature changes of the plurality of route points,
The optimization problem can be a function of a reference longitudinal displacement, of a reference lateral displacement, of a reference vehicle heading, of a reference vehicle path curvature and of a decision vector comprising the speed of variation of the curvature of the path and the distance travelled with respect to the origin and empirically determined setting parameters, and constraints on the initial longitudinal displacement, on the initial lateral displacement, on t
B · US 2024/0034311 A1 · ¶ 15
1.4a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: determine a second adjustment value for moving the plurality of route points toward a lane center of the driving lane of the vehicle,
Trincavelli, ECC, Linz (2015)” discloses a control method for centering in the lane.
B · US 2024/0034311 A1 · ¶ 7
1.5a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: adjust, based on the first adjustment value and the second adjustment value, the plurality of route points,
Adjustments can be made to one or more of the waypoint locations in order to accommodate this mode of transportation in view of a given cost function. Adjustments can be made that optimize the cost function to minimize overall costs.
A · US 7,693,652 B2 · abstract
1.6a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: generate an updated driving route comprising the adjusted plurality of route points,
Turning now to FIG. 2, there is a block diagram of a routing system 200 that facilitates adjusting waypoints to generate a more reasonable route in part by analyzing user state and inferring user intentions.
A · US 7,693,652 B2 · ¶ 33
1.7a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: output a signal indicating the updated driving route, and
1.8a memory storing at least one instruction that, when executed by the processor, is configured to cause the apparatus to: control, based on the signal, an operation of the vehicle.
The control method includes: determining data of the sensors, fusing the data of the sensors so as to determine the steering wheel angle, the vehicle speed and the vehicle heading, planning an avoidance path to avoid the target, taking the form of a Euler spiral, refining the avoidance path depending on the steering wheel angle, on the vehicle speed and on the vehicle heading and based on the solution of an optimisation problem, controlling the vehicle so as to follow the ref
B · US 2024/0034311 A1 · abstract
An empty cell means only that no passage was reproduced there.
Statements in these documents
For instance, such conventional systems rely on and in most cases require address input in order to determine a reasonable set of driving directions.
A · US 7,693,652 B2 · ¶ 1
The subject application relates to a system(s) and methodology that facilitate providing a more reasonable route given arbitrarily selected waypoints such as a start, end, and any other points in between.
A · US 7,693,652 B2 · ¶ 4
However, despite all its performance levels, this robust approach to path planning lacks flexibility because it imposes strict constraints on the paths sought (that is to say the final angle of the steering wheel is equal to 0, the integral of the steering wheel angle profile is equal to 0) and because the non-zero initial conditions of the system are disregarded.
B · US 2024/0034311 A1 · ¶ 3
That risks rendering the path tracking controller and the vehicle system unstable.
B · US 2024/0034311 A1 · ¶ 3
Cite aspatent.fyi, US 2026/0356391 A1, claim 1, first published 2026-11-20T01:05:31-05:00, https://patent.fyi/US20260356391/, SHA-256 d2c395e0e16120b77bcf936e2359661bb5d51bfd46c43dc3b2cc792d52bbcf3f