METHOD AND DEVICE FOR ESTIMATING ROAD SURFACE USING VEHICLE DETECTION, COMPUTER-READABLE NON-TRANSIENTIAL MEDIUM, AND APPARATUS

Abstract

Various aspects of the present disclosure generally relate to autonomous driving and driver assistance systems. In some aspects, a device associated with an ego vehicle may obtain a series of images that depict a reference vehicle traveling along a road segment ahead of the ego vehicle. The device may estimate, based on the series of images, a size of the reference vehicle and / or a position of the reference vehicle relative to the ego vehicle. The device may track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on the estimated size of the reference vehicle and / or the estimated position of the reference vehicle over the series of images. The device may estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle. Numerous other aspects are described.

Description

"METHOD AND DEVICE FOR ESTIMATING ROAD SURFACE USING VEHICLE DETECTION, COMPUTER-READABLE NON-TRANSIENTIAL MEDIUM, AND APPARATUS" CROSS-REFERENCE TO RELATED REQUESTS

[0001] This patent application claims priority from non-provisional US patent application No. 18 / 187,395, filed March 21, 2023, entitled ROAD SURFACE ESTIMATION USING VEHICLE DETECTION, which is hereby expressly incorporated by reference. FIELD OF DISSEMINATION

[0002] Aspects of this disclosure relate generally to autonomous driving and / or an advanced driver assistance system (ADAS) and, for example, to the estimation of road surface geometry using vehicle detection in conditions where a road surface may be difficult to estimate from direct visual observation. BACKGROUND

[0003] Autonomous driving systems are an emerging technology that allows a vehicle to operate without human intervention, following a pre-programmed route or responding to environmental conditions in real time. Autonomous driving systems generally use a combination of sensors, cameras, and software algorithms to perceive the environment and make decisions based on the perceived environment. Autonomous driving technology can be designed to create a safer, more efficient, and more convenient mode of transportation that reduces the need for human intervention. The development of autonomous driving systems was Petition 870250080995, dated 09 / 09 / 2025, page 11 / 151 2 / 56 driven by a convergence of factors (e.g., advances in sensor technology, artificial intelligence, and machine learning) that enabled vehicles to detect and process information from their surrounding environment (e.g., road conditions, traffic, and pedestrians). SUMMARY

[0004] Some aspects described in the present invention relate to a method for estimating road surface using vehicle detection. The method may include obtaining, by a device associated with an ego vehicle, a series of images depicting a reference vehicle moving along a road segment ahead of the ego vehicle. The method may include estimating, by the device based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle. The method may include tracking, by the device, a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images.The method may involve the device estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0005] Some aspects described in the present invention relate to a device for estimating road surface using vehicle detection. The device may include a memory and one or more processors coupled to the memory. The one or more processors may be configured to obtain a series of images depicting a vehicle. Petition 870250080995, dated 09 / 09 / 2025, page 12 / 151 3 / 56 Reference moving along a road segment ahead of an ego vehicle. One or more processors can be configured to estimate, based on the image series, one or more of the reference vehicle's size or position relative to the ego vehicle. One or more processors can be configured to track a reference vehicle's trajectory along the road segment ahead of the ego vehicle based on one or more of the estimated reference vehicle's size or position relative to the image series. One or more processors can be configured to estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0006] Some aspects described in the present invention relate to a non-transient, computer-readable means that stores a set of instructions. The instruction set, when executed by one or more processors of a device, can cause the device to obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle. The instruction set, when executed by one or more processors of the device, can cause the device to estimate, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle. The instruction set, when executed by one or more processors of the device, can cause the device to track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the size Petition 870250080995, dated 09 / 09 / 2025, page 13 / 151 4 / 56 estimated of the reference vehicle or the estimated position of the reference vehicle relative to the image series. The instruction set, when executed by one or more device processors, can cause the device to estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0007] Some aspects described in the present invention relate to an apparatus. The apparatus may include means for obtaining a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle. The apparatus may include means for estimating, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle. The apparatus may include means for tracking a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images. The apparatus may include means for estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0008] The aspects generally include a method, an apparatus, a system, a computer program product, a non-transient computer-readable medium, a user device, a user equipment, a wireless communication device and / or a processing system, as substantially described with reference to and illustrated by the drawings and the report. Petition 870250080995, dated 09 / 09 / 2025, page 14 / 151 5 / 56 described.

[0009] The foregoing has described in a fairly broad manner the attributes and technical advantages of the examples according to the disclosure so that the detailed description that follows may be better understood. Additional attributes and advantages will be described hereafter. The specific design and examples disclosed may be readily used as a basis for modifying or designing other structures to accomplish the same purposes as the present disclosure. Such equivalent constructions do not depart from the scope of the appended claims. The characteristics of the concepts disclosed in the present invention, both their organization and method of operation, together with associated advantages, will be better understood from the following description when considered in conjunction with the accompanying figures. Each of the figures is provided for the purpose of illustration and description, and not as a definition of the limits of the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order for the attributes mentioned above in this disclosure to be understood in detail, a more particular description, briefly summarized above, can be obtained by reference to aspects, some of which are illustrated in the accompanying drawings. It should be mentioned, however, that the accompanying drawings illustrate only certain typical aspects of this disclosure and, therefore, should not be considered limiting to its scope, as the description may include other equally effective aspects. Identical reference numbers in different drawings may identify identical or similar elements. Petition 870250080995, dated 09 / 09 / 2025, page 15 / 151 6 / 56

[0011] Figure 1 is a diagram of an example environment in which an autonomous vehicle or a vehicle equipped with an advanced driver assistance system (ADAS) may operate, according to this disclosure.

[0012] Figure 2 is a diagram of an example of an onboard system for an autonomous vehicle or a vehicle equipped with an ADAS, according to this disclosure.

[0013] Figure 3 is a diagram illustrating sample components of one or more devices shown in Figure 1 and / or Figure 2, according to this disclosure.

[0014] Figures 4A and 4B are diagrams illustrating examples associated with estimating a road surface using vehicle detection, according to this disclosure.

[0015] Figure 5 is a flowchart of an example process associated with estimating a road surface using vehicle detection, according to this disclosure. DETAILED DESCRIPTION

[0016] Several aspects of the disclosure are described more fully hereafter, with reference to the accompanying drawings. This disclosure can, however, be incorporated in many different forms and should not be interpreted as limited to any specific structure or function presented throughout this disclosure. Rather, these aspects are provided so that this disclosure is thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. Those skilled in the art will recognize that the scope of the disclosure is intended to cover any aspect of the disclosure disclosed in the present invention, whether this Petition 870250080995, dated 09 / 09 / 2025, p. 16 / 151 7 / 56 implemented independently or in combination with any other aspect of the disclosure. For example, an apparatus may be implemented, or a method may be practiced using any number of the aspects set forth in the present invention. Furthermore, the scope of the disclosure is intended to cover such an apparatus or method that is practiced using another structure, functionality, or structure and functionality in addition to or different from the various aspects of the disclosure set forth in the present invention. It should be understood that any aspect of the disclosure disclosed in this invention may be incorporated by one or more elements of a claim.

[0017] Estimating a road surface in terms of trajectory, shape, alignment, profile, and / or other parameters is used in many autonomous driving and / or advanced driver assistance system (ADAS) applications to enable the autonomous driving and / or ADAS application to make an informed decision about how to navigate the road safely and efficiently. For example, estimating a road surface can be critically important for maintaining control with respect to an autonomous vehicle or a vehicle equipped with an ADAS because different road surfaces (e.g., concrete, asphalt, and gravel) have different levels of traction and / or may require different handling characteristics.Furthermore, the alignment (e.g., tangents and horizontal curves) and profile (e.g., vertical aspect, including crest and camber curves and connecting straight slope lines) of a road surface play an important role in vehicle control because changes in road surface geometry can affect vehicle stability and handling. In another example, many... Petition 870250080995, dated 09 / 09 / 2025, page 17 / 151 8 / 56 Autonomous driving and ADAS applications rely on road surface information to keep a vehicle within a designated lane (e.g., by estimating a road alignment and profile, an autonomous vehicle or ADAS application can make corrections to maintain a consistent path, even in cases where there are changes in the road surface). Furthermore, road surface information can be used for obstacle avoidance (e.g., by anticipating changes in road geometry and adjusting a vehicle's trajectory accordingly to avoid potential obstacles) and / or other suitable applications (e.g., route planning).

[0018] In general, information related to road surface geometry is typically estimated using camera-based vision systems. For example, monocular camera-based vision techniques may use a single camera to capture images of a roadway, and the images can be analyzed using computer vision algorithms to estimate a vehicle's position relative to the roadway, as well as the shape of the roadway surface, the location of lane markers, and / or other relevant information about the roadway surface. In another example, stereo camera-based vision uses two cameras to capture images of the roadway, enabling depth perception and the creation of three-dimensional (3D) models of the roadway surface that can then be used to estimate the height and slope of the roadway, as well as the position of objects on or near the roadway. In yet another example, light detection and ranging (lidar) uses laser light to create a 3D map of the roadway surface.Consequently, existing techniques that use camera-based vision systems to estimate a geometry of... Petition 870250080995, dated 09 / 09 / 2025, page 18 / 151 9 / 56 Road surface studies are generally focused on capturing and analyzing images of the road surface. However, there are several situations where an accurate estimate of the road surface can be difficult to achieve through direct visual observation of the road surface. For example, the road surface may not be visible in images that are captured in adverse weather conditions or in adverse light conditions (e.g., fog, rain, snow, glare, and / or night), when the road surface is occluded (e.g., by other vehicles, vegetation, and / or stationary or moving objects), and / or when there are undulating roads (e.g., having an ascending and descending shape that causes parts of the road to be invisible).

[0019] In some aspects described in the present invention, a system or onboard device associated with an ego vehicle can capture or otherwise obtain a series of images depicting one or more reference vehicles moving along a road segment ahead of the ego vehicle and analyze the size and / or position of the reference vehicles in the image series to estimate the surface geometry of the road segment ahead of the ego vehicle. For example, as the reference vehicles are moving along the road segment ahead of the ego vehicle, the trajectory of the reference vehicles can be tracked relative to the image series and used to estimate the surface geometry of the next road segment ahead of the ego vehicle. Consequently, since vehicles can be more visible than the road surface under adverse conditions and vehicles are rigid objects with a fixed size that does not change over time, an accurate size estimate Petition 870250080995, dated 09 / 09 / 2025, p. 19 / 151 A 10 / 56 measurement for a vehicle at a point in time will continue to be accurate as time progresses and can therefore serve as a scale reference. In this way, by tracking the size and relative position of a reference vehicle over time (e.g., across a series of images), a trajectory traveled by the reference vehicle can be traced and used to estimate the surface geometry of the road segment traveled by the reference vehicle. Furthermore, using visual odometry and / or vehicle sensors (e.g., inertial measurement units and / or positioning systems, among other examples), an accurate estimate of the road surface traveled by the ego vehicle can be determined.Consequently, by comparing the accurate trajectory of the ego vehicle relative to a traveled portion of a road segment with the trajectory of the reference vehicle that was estimated for the traveled portion of the road segment, the size estimate of the reference vehicle can be refined and used to improve the road surface estimate for an untraveled portion of the road segment.

[0020] Figure 1 is a diagram of an example environment 100 in which an autonomous vehicle or a vehicle equipped with an advanced driver assistance system (ADAS) may operate, according to this disclosure. As shown in Figure 1, environment 100 may include, for example, a vehicle 110, an on-board system 120 of the vehicle 110, a remote device 130, a network node 150, and a network 160. The devices in environment 100 may interconnect via wired connections, wireless connections, or a combination of wired and wireless connections. As further shown in Figure 1, environment 100 may include one or more objects Petition 870250080995, dated 09 / 09 / 2025, page 20 / 151 11 / 56 140 that the vehicle 110 is configured to detect (for example, using the on-board system 120).

[0021] In some respects, vehicle 110 may include any mobile form of transport capable of carrying one or more human occupants and / or cargo and powered by any suitable energy source. For example, vehicle 110 may include a land vehicle (e.g., a car, a truck, a van, or a train), an aircraft (e.g., an unmanned aerial vehicle), and / or a vessel. In the example depicted in Figure 1, vehicle 110 is a land vehicle and is shown as a car. Furthermore, vehicle 110 is an autonomous vehicle in the example in Figure 1. For example, an autonomous vehicle (AV) is a vehicle with a processor, programming instructions, and transmission components that are controllable by the processor without requiring a human operator.An autonomous vehicle can be fully autonomous, meaning the autonomous vehicle does not require a human operator for most or all driving conditions and functions, or it can be semi-autonomous, meaning a human operator may be required under certain conditions or for certain operations, or a human operator may override the autonomous vehicle's autonomous system and take control of the autonomous vehicle. Additionally or alternatively, the 110 vehicle may be equipped with an ADAS that supports one or more safety attributes and / or technologies to assist drivers in avoiding collisions and / or accidents (e.g., adaptive cruise control, lane departure warning, automatic emergency braking) or otherwise make driving the 110 vehicle safer and / or more efficient. Petition 870250080995, dated 09 / 09 / 2025, page 21 / 151 12 / 56

[0022] As shown in Figure 1, the vehicle 110 may include an on-board system 120 that is integrated and / or coupled to the vehicle 110. In general, the on-board system 120 may be used to control the vehicle 110, to detect information about the vehicle 110 and / or an environment in which the vehicle 110 operates, to detect one or more objects 140 in the vicinity of the vehicle, to provide output or receive input from an occupant of the vehicle 110 and / or to communicate with one or more remote devices of the vehicle 110, such as another vehicle and / or the remote device 130. Consequently, as described in the present invention, the vehicle 110 may be an ego vehicle, which refers to the vehicle in question that is using autonomous driving technology, an ADAS and / or one or more sensors (e.g., cameras, lidars and radars) to perceive a surrounding environment and make decisions related to a trajectory, speed and / or actions of the vehicle 110 on the road.The onboard system 120 is described in more detail below in conjunction with Figure 2.

[0023] In some respects, vehicle 110 can move along a roadway in a semi-autonomous or autonomous manner. Vehicle 110 can be configured to detect objects 140 in the vicinity of vehicle 110. An object 140 may include, for example, another vehicle (e.g., an autonomous vehicle or a non-autonomous vehicle that requires a human operator for all or almost all driving conditions and functions), a cyclist (e.g., a cyclist on a bicycle, electric scooter, or motorcycle), a pedestrian, a roadway attribute (e.g., a highway boundary, a lane marker, a sidewalk, a median, a guardrail, a barricade, a sign, a traffic sign, a Petition 870250080995, dated 09 / 09 / 2025, page 22 / 151 13 / 56 railway crossing or a cycle path) and / or another object that may be on or near a roadway, such as a tree or an animal. In some respects, to detect objects 140, the vehicle 110 may be equipped with a camera-based vision system and / or one or more sensors, such as a lidar system. In some respects, the camera-based vision system and / or one or more sensors may be included in another system other than the vehicle 110, such as a robot, a satellite and / or a traffic light.

[0024] In some respects, one or more sensors may provide object detection data, such as information about a detected object 140 (e.g., information about a distance to the object 140, a speed of the object 140 and / or a direction of movement of the object 140) to one or more other onboard system components 120. Additionally or alternatively, the vehicle 110 may transmit the object detection data to the remote device 130 (e.g., a server, a cloud computing system and / or a database) via the network 160 (e.g., via the network node 150). The remote device 130 may be configured to process the object detection data and / or to transmit a result of the processing of the object detection data to the vehicle 110 via the network 160 (e.g., via the network node 150).

[0025] In some respects, network node 150 includes one or more devices configured to receive, generate, store, process and / or provide information relating to one or more aspects described in the present invention. For example, network node 150 may include a base station (a nodeB, a gNB and / or a 5G B (NB) node, among other examples), Petition 870250080995, dated 09 / 09 / 2025, page 23 / 151 14 / 56 a user equipment (UE), a relay device, a network controller, an access point, a transmission reception point (TRP), an apparatus, a device, a computing system and / or other suitable processing entity configured to perform one or more aspects described in the present invention. For example, in some aspects, the network node 150 may include an aggregated base station and / or one or more components of a disaggregated base station (e.g., a central unit, a distributed unit and / or a radio unit) that enables the onboard system 120 to communicate through the network 160 (e.g., to invoke or otherwise utilize processing capabilities associated with the remote device 130).

[0026] The 160 network includes one or more wired and / or wireless networks. For example, the 160 network may include a cellular network (e.g., a long-term evolution (LTE) network, a code division multiple access (CDMA) network, a third-generation (3G) network, a fourth-generation (4G) network, a fifth-generation (5G) network, another type of next-generation network and / or similar), a public land mobile network (PLMN), a local area network (LAN), a wide area network (WAN), a metropolitan area network (MAN), a telephone network (e.g., a public switched telephone network (PSTN)), a private network, an ad hoc network, an intranet, a fiber optic-based network, a network of Petition 870250080995, dated 09 / 09 / 2025, page 24 / 151 15 / 56 cloud computing or similar, and / or a combination of these and other types of networks. In some respects, the 160 network enables communication between devices in the 100 environment.

[0027] In some respects, as described in the present invention, the onboard system 120 can be configured to obtain a series of images depicting a reference vehicle moving along a road segment ahead of vehicle 110; estimate, based on the series of images, one or more of the reference vehicle's size or position relative to vehicle 110; track a trajectory of the reference vehicle along the road segment ahead of vehicle 110 based on one or more of the estimated reference vehicle's size or position relative to the series of images; and estimate a surface geometry associated with the road segment ahead of vehicle 110 based on the tracked trajectory of the reference vehicle.

[0028] As indicated above, Figure 1 is provided as an example. Other examples may differ from those described in relation to Figure 1. The number and arrangement of devices shown in Figure 1 are provided as an example. In practice, there may be additional devices, fewer devices, different devices, or devices arranged differentially in relation to those shown in Figure 1. Furthermore, two or more devices shown in Figure 1 may be implemented in a single device, or a single device shown in Figure 1 may be implemented as multiple distributed devices. Additionally or alternatively, a set of devices (e.g., Petition 870250080995, dated 09 / 09 / 2025, page 25 / 151 16 / 56 one or more devices) shown in Figure 1 can perform one or more functions described as being performed by another set of devices shown in Figure 1.

[0029] Figure 2 is a diagram of an example of an on-board system 200 of an autonomous vehicle or a vehicle equipped with an ADAS, according to the present disclosure. In some respects, the on-board system 200 may correspond to the on-board system 120 included in the vehicle 110, as described above in conjunction with Figure 1. As shown in Figure 2, the on-board system 200 may include one or more of the components illustrated 202 to 256. The on-board system 200 may include, for example, a power subsystem 202, a sensor subsystem 204, a control subsystem 206 and / or an on-board device 208. The components of the on-board system 200 may communicate via a bus (e.g., one or more wired and / or wireless connections), such as a controller area network (CAN) bus.

[0030] The energy subsystem 202 can be configured to generate mechanical energy for vehicle 110 to move vehicle 110. For example, the energy subsystem 202 may include an engine that converts fuel into mechanical energy (e.g., via combustion) and / or an engine that converts electrical energy into mechanical energy.

[0031] Sensor subsystem 204 may include one or more sensors configured to detect vehicle operating parameters 110 and / or environmental conditions in an environment in which vehicle 110 operates (e.g., surrounding vehicle 110). For example, sensor subsystem 204 may include an engine temperature sensor 210, a battery voltage sensor 212, a sensor of Petition 870250080995, dated 09 / 09 / 2025, page 26 / 151 17 / 56 engine revolutions per minute (RPM) 214, a throttle position sensor 216, a battery sensor 218 (e.g., to measure current, voltage and / or temperature of a battery), an engine current sensor 220, an engine voltage sensor 222, an engine position sensor 224 (for example, a resolver and / or encoder), a motion sensor 226 (for example, an accelerometer, gyroscope and / or inertial measuring unit), a speed sensor 228, an odometer sensor 230, a clock 232, a position sensor 234 (for example, a global navigation satellite system (GNSS) sensor and / or a global positioning system (GPS) sensor), one or more cameras 236, a lidar system 238, one or more other ranging systems 240 (for example, a radar system and / or a sonar system) and / or an environmental sensor 242 (for example, a precipitation sensor and / or an ambient temperature sensor).

[0032] The control subsystem 206 may include one or more controllers configured to control the operation of vehicle 110. For example, the control subsystem 206 may include a brake controller 244 to control the braking of vehicle 110, a steering controller 246 to control the steering and / or direction of vehicle 110, a throttle controller 248 and / or a speed controller 250 to control the speed and / or acceleration of vehicle 110, a gear controller 252 to control the gear shifting of vehicle 110, a routing controller 254 to control the navigation and / or routing of vehicle 110 (e.g., using map data) and / or an auxiliary device controller 256 to Petition 870250080995, dated 09 / 09 / 2025, page 27 / 151 18 / 56 control one or more auxiliary devices associated with vehicle 110, such as a test device, an auxiliary sensor and / or a mobile device carried by vehicle 110.

[0033] The on-board device 208 can be configured to receive sensor data from one or more sensors included in the sensor subsystem 204 and / or to provide commands to one or more controllers included in the control subsystem 206. For example, the on-board device 208 can control the operation of vehicle 110 by providing a command to a controller included in the control subsystem 206 based on sensor data received from a sensor included in the sensor subsystem 204. In some respects, the on-board device 208 can be configured to process sensor data to generate a command. The on-board device 208 may include memory, one or more processors, an input component, an output component, and / or a communication component, as described elsewhere in the present invention.

[0034] As an example, the onboard device 208 can receive navigation data, such as information associated with a navigation route from a vehicle's starting location 110 to a destination location for vehicle 110. In some respects, the navigation data is accessed and / or generated by the routing controller 254. For example, the routing controller 254 can access map data and identify possible routes and / or road segments that vehicle 110 can travel to move from the starting location to the destination location. In some respects, the routing controller 254 can identify a preferred route, such as scoring multiple possible routes, applying one or more routing techniques (e.g., Petition 870250080995, dated 09 / 09 / 2025, page 28 / 151 19 / 56 minimum Euclidean distance, Dijkstra's algorithm and / or Bellman-Ford algorithm), accounting for traffic data and / or receiving a user selection of a route, among other examples. The onboard device 208 can use the navigation data to control the operation of the vehicle 110. As the vehicle moves along the route, the onboard device 208 can receive sensor data from various sensors in the sensor subsystem 204. For example, the position sensor 234 can provide geographic location information to the onboard device 208, which can then access a map associated with the geographic location information to determine known fixed attributes associated with the geographic location, such as streets, buildings, stop signs and / or traffic signs, which can be used to control the operation of the vehicle 110.

[0035] In some respects, the onboard device 208 can receive one or more images captured by one or more cameras 236, can analyze one or more images (e.g., to detect object data), and can control the operation of vehicle 110 based on the analysis of the images (e.g., to avoid detected objects). For example, the onboard device 208 can obtain, from camera(s) 236, a series of images depicting a reference vehicle moving along a road segment ahead of vehicle 110, and the onboard device 208 can analyze the series of images to estimate a size of the reference vehicle and / or a position of the reference vehicle relative to vehicle 110. The onboard device 208 can track a trajectory of the reference vehicle along the road segment ahead of vehicle 110 based on the estimated size of the reference vehicle. Petition 870250080995, dated 09 / 09 / 2025, page 29 / 151 20 / 56 reference and / or the estimated position of the reference vehicle relative to the image series and can estimate a surface geometry associated with the road segment ahead of vehicle 110 based on the tracked trajectory of the reference vehicle. Consequently, the onboard system 208 can generate one or more control signals (e.g., to control vehicle 110, stay within a designated lane, avoid an obstacle, and / or plan a route) based on the estimated surface geometry associated with the road segment ahead of vehicle 110.

[0036] In some respects, the onboard device 208 may receive object data associated with one or more objects detected in the vicinity of the vehicle 110 and / or may generate object data based on sensor data. The object data may indicate the presence or absence of an object, an object location, a distance between the object and the vehicle 110, an object speed, an object direction of movement, an object acceleration, an object trajectory (e.g., a heading), an object shape, an object size, an object projection area, and / or an object type (e.g., a vehicle, a pedestrian, a cyclist, a stationary object, or a moving object). The object data may be detected, for example, by one or more cameras 236 (e.g., as image data), the lidar system 238 (e.g., as lidar data), and / or one or more other range systems 240 (e.g., as radar data or sonar data).The on-board device 208 can process object data to detect objects in the vicinity of vehicle 110 and / or to control the operation of vehicle 110 based on object data (e.g., to avoid detected objects). Petition 870250080995, dated 09 / 09 / 2025, page 30 / 151 21 / 56

[0037] In some respects, the onboard device 208 can use object data (e.g., current object data) to predict future object data for one or more objects. For example, the onboard device 208 can predict a future location of an object, a future distance between the object and vehicle 110, a future speed of the object, a future direction of movement of the object, a future acceleration of the object, and / or a future trajectory (e.g., a future heading) of the object. For example, if an object is a vehicle and the map data indicates that the vehicle is at an intersection, then the onboard device 208 can predict whether the object is likely to move in a straight line or turn. As another example, if the sensor data and / or the map data indicate that the intersection does not have a traffic light, then the onboard device 208 can predict whether the object will stop before entering the intersection.

[0038] The onboard device 208 can generate a motion plan for vehicle 110 based on sensor data, navigation data, and / or object data (e.g., current object data and / or future object data). For example, based on current object locations and / or predicted future object locations, the onboard device 208 can generate a motion plan to move vehicle 110 along a surface and avoid collision with other objects. In some respects, the motion plan may include, at one or more points in time, a vehicle 110 speed, a vehicle 110 direction, and / or a vehicle 110 acceleration. Additionally or alternatively, the motion plan may indicate one or more actions in relation to a detected object, such as overtaking the object, yielding to the object, passing the Petition 870250080995, dated 09 / 09 / 2025, page 31 / 151 22 / 56 object or similar. The onboard device 208 can generate one or more commands or instructions based on the motion plan and can provide these commands to one or more controllers associated with the control subsystem 206 for execution.

[0039] As indicated above, Figure 2 is provided as an example. Other examples may differ from that described in relation to Figure 2. The number and arrangement of components shown in Figure 2 are provided as an example. In practice, there may be additional components, a smaller number of components, different components, or components arranged differently from those shown in Figure 2. Furthermore, two or more components shown in Figure 2 may be implemented as single components, or a single component shown in Figure 2 may be implemented as multiple distributed components. Additionally or alternatively, a set of components (e.g., one or more components) shown in Figure 2 may perform one or more functions described as being performed by another set of components shown in Figure 2.For example, although some components of Figure 2 are primarily associated with land vehicles, other types of vehicles are within the scope of the disclosure.

[0040] Figure 3 is a diagram illustrating example components of a device 300, according to the present disclosure. The device 300 may correspond to the onboard system 120, the remote device 130 or the network node 150 depicted in Figure 1, the onboard system 200 or the onboard device 208 depicted in Figure 2 and / or any other device, system, subsystem or component described in the present invention. In some Petition 870250080995, dated 09 / 09 / 2025, page 32 / 151 23 / 56 aspects, the onboard system 120, the remote device 130, the network node 150, the onboard system 200, the onboard device 208 and / or other devices, systems, subsystems or components described in the present invention may include one or more devices 300 and / or one or more components of the device 300. As shown in Figure 3, the device 300 may include a bus 305, a processor 310, a memory 315, an input component 320, an output component 325, a communication component 330 and / or an estimation component 335.

[0041] Bus 305 may include one or more components that enable wired and / or wireless communication between the components of device 300. Bus 305 may couple two or more components of Figure 3, such as by means of operational coupling, communicative coupling, electronic coupling, and / or electrical coupling. For example, bus 305 may include an electrical connection (e.g., a wire, a trace, and / or a conductor) and / or a wireless bus. Processor 310 may include a central processing unit, a graphics processing unit, a microprocessor, a controller, a microcontroller, a digital signal processor, a field-programmable gate array, an application-specific integrated circuit, and / or other type of processing component. Processor 310 may be implemented in hardware, firmware, or a combination of hardware and software.In some respects, the 310 processor may include one or more processors capable of being programmed to perform one or more operations or processes described elsewhere in the present invention. Petition 870250080995, dated 09 / 09 / 2025, page 33 / 151 24 / 56

[0042] Memory 315 may include volatile and / or non-volatile memory. For example, memory 315 may include random access memory (RAM), read-only memory (ROM), a hard disk, and / or other types of memory (e.g., flash memory, magnetic memory, and / or optical memory). Memory 315 may include internal memory (e.g., RAM, ROM, or a hard disk) and / or removable memory (e.g., removable via a universal serial bus connection). Memory 315 may be a non-transient, computer-readable medium. Memory 315 may store information, one or more instructions, and / or software (e.g., one or more software applications) related to the operation of device 300. In some respects, memory 315 may include one or more memories that are coupled (e.g., communicatively coupled) to one or more processors (e.g., processor 310), such as via bus 305.Communicative coupling between a 310 processor and a 315 memory can enable the 310 processor to read and / or process information stored in memory 315 and / or store information in memory 315.

[0043] Input component 320 can enable device 300 to receive input, such as user input and / or detected input. For example, input component 320 can include a touch screen, a keyboard, a numeric keypad, a mouse, a button, a microphone, a switch, a sensor, a global positioning system sensor, a global navigation satellite system sensor, an accelerometer, a gyroscope, and / or an actuator. Output component 325 can enable device 300 to provide Petition 870250080995, dated 09 / 09 / 2025, page 34 / 151 25 / 56 output, such as through a screen, a speaker, and / or a light-emitting diode. The communication component 330 can enable the device 300 to communicate with other devices through a wired and / or wireless connection. For example, the communication component 330 may include a receiver, a transmitter, a transceiver, a modem, a network interface card, and / or an antenna.

[0044] The estimation component 335 can obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; estimate, based on the image series, a size of the reference vehicle and / or a position of the reference vehicle relative to the ego vehicle; track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on the estimated size of the reference vehicle and / or the estimated position of the reference vehicle relative to the image series; and estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0045] The device 300 can perform one or more operations or processes described in the present invention. For example, a non-transient computer-readable medium (e.g., memory 315) can store a set of instructions (e.g., one or more instructions or code) for execution by the processor 310. The processor 310 can execute the set of instructions to perform one or more operations or processes described in the present invention. In some respects, the execution of the set of instructions by one or more processors 310 causes the one or more processors 320 and / or the device 300 to perform one or Petition 870250080995, dated 09 / 09 / 2025, page 35 / 151 26 / 56 more operations or processes described in the present invention. In some respects, the permanently connected circuitry can be used instead of, or in combination with, the instructions to perform one or more operations or processes described in the present invention. Additionally or alternatively, the 310 processor can be configured to perform one or more operations or processes described in the present invention. Thus, the aspects described in the present invention are not limited to any specific combination of hardware and software circuitry.

[0046] In some respects, the 300 device may include means for obtaining a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; means for estimating, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; means for tracking a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and / or means for estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.In some respects, the means for the device 300 to perform processes and / or operations described in the present invention may include one or more components of the device 300 described together in Figure 3, such as bus 305, processor 310, memory 315, input component 320, output component 325, communication component 330. Petition 870250080995, dated 09 / 09 / 2025, page 36 / 151 27 / 56 and / or estimation component 335. Additionally or alternatively, the means for the device 300 to perform processes and / or operations described in the present invention may include one or more components of the onboard system 200 described in conjunction with Figure 2, such as the sensor subsystem 204, the control subsystem 206 and / or the onboard device 208, among other examples.

[0047] The number and arrangement of components shown in Figure 3 are provided as an example. Device 300 may include additional components, a smaller number of components, different components, or components arranged differently from those shown in Figure 3. Additionally or alternatively, a set of components (e.g., one or more components) of device 300 may perform one or more functions described as being performed by another set of components of device 300.

[0048] Figures 4A and 4B are diagrams illustrating examples 400 associated with estimating a road surface using vehicle detection, according to the present disclosure. As shown in Figures 4A and 4B, examples 400 include an ego vehicle (e.g., vehicle 110) equipped with an onboard system (e.g., onboard system 120, onboard system 200, and / or onboard device 208) that supports autonomous driving and / or ADAS technology. As further shown in Figures 4A and 4B, examples 400 include a reference vehicle traveling on a road segment ahead of the ego vehicle. Consequently, as described in the present invention, examples 400 refer to techniques that can be implemented in the ego vehicle's onboard system to estimate the surface geometry of a road surface. Petition 870250080995, dated 09 / 09 / 2025, page 37 / 151 28 / 56 of track ahead of the ego vehicle based on detections from the reference vehicle that are tracked over time.

[0049] For example, as shown in Figure 4A, and by reference number 405, the onboard system associated with the ego vehicle can obtain a series of images depicting the reference vehicle while the reference vehicle is moving along a road segment ahead of the ego vehicle. For example, in some aspects, the onboard system may include or receive input from a camera system that can capture the series of images depicting the reference vehicle moving along the road segment ahead of the ego vehicle, and the image series can then be analyzed to estimate a surface geometry associated with the road segment ahead of the ego vehicle, using techniques described in more detail in the present invention. In some aspects, the camera system may include a monocular camera system that uses a single camera to capture images of the ego vehicle's surroundings. Alternatively, the camera system may include a stereo-based vision system that uses two cameras mounted on either side of the ego vehicle to capture images from slightly different angles. In some respects, the camera(s) used to capture the series of images depicting the reference vehicle may include one or more forward-facing cameras (e.g., mounted on the front of the ego vehicle), one or more surround-view cameras (e.g., using multiple cameras positioned around the ego vehicle to provide a 360-degree view of the ego vehicle's surroundings), and / or one or more infrared cameras that may be used for... Petition 870250080995, dated 09 / 09 / 2025, page 38 / 151 29 / 56 detect objects in low light conditions.

[0050] As further shown in Figure 4A, and by reference number 410, the onboard system can estimate a size of the reference vehicle based on the series of images, and the onboard system can further estimate a position of the reference vehicle relative to a position of the ego vehicle in each image. For example, as described in the present invention, the reference vehicle is a rigid object that has a fixed size, so the estimated size of the reference vehicle can serve as a scale reference to estimate a distance between the ego vehicle and the reference vehicle in each image of the reference vehicle. For example, in cases where the reference vehicle appears smaller in a first image than in a second image, the reference vehicle is closer to the ego vehicle in the second image.Consequently, in some respects, the onboard system can estimate a measurement related to the size of the reference vehicle in order to determine a scale reference for estimating the position of the reference vehicle relative to a position of the ego vehicle in each image. For example, the onboard system can estimate a height of the reference vehicle, a width of the reference vehicle, or a dimension associated with any other suitable attribute of the reference vehicle (e.g., a wheel size, a license plate size, a taillight size, and / or a rear window size, among other examples). In some respects, in cases where the measurement depends on a design of the reference vehicle, the onboard system can use computer vision techniques to determine one or more parameters that indicate the design of the reference vehicle (e.g., Petition 870250080995, dated 09 / 09 / 2025, page 39 / 151 30 / 56 a body style, make and model, or similar), which can be used to determine the appropriate size estimate (for example, a height estimate for a bus or semi-trailer truck may be significantly larger than a height estimate for a sedan or compact car). Consequently, in some respects, the measurement related to the size of the reference vehicle can be estimated using a machine learning model (for example, a neural network or other model that implements computer vision techniques or other techniques suitable for estimating the size of a visual attribute associated with the reference vehicle). In any case, the onboard system can use the estimated measurement related to the size of the reference vehicle to estimate the position of the reference vehicle relative to the ego vehicle in relation to the image series.

[0051] For example, in some respects, the ego vehicle may use one or more position sensors, motion sensors, and / or other suitable sensors to determine an accurate position of the ego vehicle at the respective points in time when the series of images depicting the reference vehicle is captured. Furthermore, based on the estimated measurement related to the vehicle's size and a corresponding attribute size of the vehicle that is measured in pixels in each image, the onboard system can determine a distance between the ego vehicle and the reference vehicle in each image. For example, as described above, the estimated measurement related to the vehicle's size may generally correspond to any suitable well-defined visual attribute of the reference vehicle, such as a height, a width, a wheel size, a license plate size, a size Petition 870250080995, dated 09 / 09 / 2025, page 40 / 151 31 / 56 of a taillight or a rear window size, among other examples. Consequently, the size of the visual attribute can be estimated (e.g., in meters, feet, or another suitable metric), and the corresponding visual attribute can also be measured in pixels in each image. In some respects, the estimated size of the visual attribute and one or more intrinsic parameters related to the camera used to capture the images of the reference vehicle can be used to determine the distance between the ego vehicle and the reference vehicle in each image based on the number of pixels occupied by the visual attribute in each image. For example, in cases where a pinhole camera model is used to capture the series of images, the distance between the ego vehicle and the reference vehicle in each image can be calculated as follows: Equation 1 where is the distance between the reference vehicle in a particular image, is ego and the vehicle at a camera focal length, is the estimated (or presumed) measurement of the visual attribute related to the size of the reference vehicle (e.g., a height in meters or feet), and is the number of pixels corresponding to the visual attribute of the reference vehicle in the image (e.g., a height in pixels). In this example, the camera focal length is the intrinsic camera parameter used to transform a pixel size of the visual attribute in an image into the distance between the ego vehicle and the reference vehicle in the image. Additionally or alternatively, depending on the camera model and / or the Petition 870250080995, dated 09 / 09 / 2025, page 41 / 151 32 / 56 camera configuration used to capture the image series, other intrinsic parameters can be used to transform between pixel size and distance (e.g., optical center, aperture, field of view, resolution, and / or distortion characteristics). Furthermore, in some respects, a function used to transform a pixel size of a visual attribute of the reference vehicle into a distance between the ego vehicle and the reference vehicle may incorporate a noise model. For example, the noise model might define one or more parameters that relate to a size-dependent factor or bias associated with pixel measurements in an image (e.g., a size estimate associated with a pixel measurement for a reference vehicle that is at a distant distance ahead of the ego vehicle might be associated with a large error, which can be accounted for in the noise model).Alternatively, the function used to transform between pixel size and distance can incorporate parameters related to the estimated properties of the reference vehicle, such as a pose (e.g., a location and orientation associated with the reference vehicle). Alternatively, the distance between the ego vehicle and the reference vehicle in each image can be estimated using a machine learning model.

[0052] As further shown in Figure 4A, and by reference number 415, the onboard system can then estimate a surface geometry associated with the road segment ahead of the ego vehicle based on a reference vehicle trajectory that is tracked relative to the image series. For example, in some respects, Petition 870250080995, dated 09 / 09 / 2025, page 42 / 151 33 / 56 The onboard system can determine the distance between the ego vehicle and the reference vehicle in each image, as described above, and / or can determine other properties of the tracked reference vehicle, such as pose. Consequently, the estimated size of the reference vehicle can be used to estimate the position and / or orientation of the reference vehicle relative to the ego vehicle over time based on the size and position of the reference vehicle in each image (e.g., in pixels), which can indicate a trajectory of the reference vehicle along the road segment ahead of the ego vehicle. In this way, by tracking the detections of the reference vehicle over time, the estimated trajectory traveled by the reference vehicle can be tracked and used to estimate parameters related to the surface geometry of the road segment, such as an alignment and / or profile of the road segment.For example, the alignment of a road segment can refer to the route associated with the road segment, which is defined as a series of horizontal tangents (e.g., straight sections of road) and circular horizontal curves (e.g., defined by a radius, or tightening, and a deflection angle, or extension) that connect the horizontal tangents. Additionally, the profile of a road segment can be defined as a series of road slopes, or degrees, that are connected by parabolic vertical curves used to provide a gradual change from one road slope to another, so that vehicles can smoothly navigate changes in degree during travel.

[0053] Consequently, the estimated surface geometry of the track segment can then be used Petition 870250080995, dated 09 / 09 / 2025, page 43 / 151 34 / 56 in any suitable autonomous driving or ADAS application, such as vehicle control and / or lane keeping. Furthermore, in some respects, the reference vehicle's trajectory can be used to estimate the surface geometry of the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle having similar trajectories (e.g., traveling along the same road segment in the same direction, or in the same lane when there is significant variation in road alignment and / or road profile between different lanes). Additionally, in cases where the trajectories of multiple reference vehicles are tracked, the estimated surface geometry can be based on an aggregation of the trajectories that are estimated individually for each reference vehicle.Furthermore, in some respects, the estimated surface geometry of the road segment can be used to track trajectories of one or more objects that are depicted in the image series (e.g., the estimated road surface can be used to improve the estimated positions of objects over time), which can be used to improve autonomous driving (e.g., obstacle avoidance) and / or ADAS functionality (e.g., adaptive cruise control).

[0054] As further shown in Figure 4A, and by reference number 420, the onboard system can track a real trajectory traveled by the ego vehicle. For example, in some aspects, the ego vehicle may include inertial measurement units, positioning sensors, motion sensors and / or other suitable sensors and / or may support visual odometry techniques that may be Petition 870250080995, dated 09 / 09 / 2025, page 44 / 151 35 / 56 are used to accurately estimate the surface geometry of the path traveled by the ego vehicle. Consequently, measurements indicating the ego vehicle's movement can be used to determine the ego vehicle's actual trajectory relative to a portion of the road segment, which can be compared to the trajectory of the reference vehicle that was estimated for the same portion of the road segment to improve the estimated surface geometry for an untraveled portion of the road segment. For example, as shown by reference number 425, the estimated size of the reference vehicle (e.g., the estimated size of a visual attribute associated with the reference vehicle, such as height, width, wheel size, or the like) can be updated based on the actual trajectory traveled by the ego vehicle along a portion of the road segment for which the reference vehicle's trajectory was estimated.For example, as described in more detail below with reference to Figure 4B, the estimated size of the reference vehicle can be updated by determining an updated value that minimizes an error metric between the actual trajectory of the ego vehicle and the estimated trajectory of the reference vehicle along a specific portion of the road segment. As further shown in Figure 4A, and by reference number 430, the updated size estimate can then be used to improve or update the estimated trajectory of the reference vehicle, and therefore the estimated surface geometry, along an untraveled portion of the road segment ahead of the ego vehicle. Furthermore, in cases where multiple reference vehicles are tracked, the estimated surface geometry can be updated. Petition 870250080995, dated 09 / 09 / 2025, page 45 / 151 36 / 56 updating the individual trajectory estimates for each reference vehicle and aggregating the individual trajectory estimates (e.g., each new frame or image provides an update to the size estimate for each reference vehicle that is tracked, as well as the estimated surface geometry of the road segment).

[0055] For example, with reference to Figure 4B, reference number 435 indicates a real surface geometry of a road segment that is traversed by the ego vehicle, and reference number 440 depicts an estimated location of a reference vehicle when the reference vehicle was first detected (e.g., initially). Consequently, in the example depicted in Figure 4B, the ego vehicle can estimate (e.g., using the onboard system 120) the size of the reference vehicle when the reference vehicle is first detected and can then track the position and / or pose of the reference vehicle relative to the ego vehicle over time based on a series of subsequent images depicting the reference vehicle moving along the road segment ahead of the ego vehicle (e.g., based on the size of the reference vehicle in each image).In some respects, as described in the present invention, the estimated size of the reference vehicle and the position and / or pose of the reference vehicle relative to the ego vehicle can be used to track an estimated trajectory of the reference vehicle over time, from the first detection or initial detection of the reference vehicle. For example, the reference number 445 indicates the estimated surface geometry of the road segment traveled by the ego vehicle. Petition 870250080995, dated 09 / 09 / 2025, page 46 / 151 37 / 56 reference, which begins at the estimated position of the reference vehicle when the reference vehicle was first detected. Consequently, after the ego vehicle has traversed a portion of the road segment that overlaps with a portion of the road segment for which surface geometry was estimated based on the tracked trajectory of the reference vehicle, the onboard system associated with the ego vehicle may update or otherwise improve the size estimates of the reference vehicle to update or otherwise improve an estimated surface geometry for a next (untraveled) portion of the road segment that is covered by the historical tracked trajectory for the reference vehicle.For example, in Figure 4B, the reference number 450 depicts an actual road geometry for the road segment along which the reference vehicle and the ego vehicle are traveling, which deviates from the road geometry estimated based on the tracked trajectory of the reference vehicle.

[0056] Consequently, as shown by reference number 455, the onboard system can calculate an updated size estimate for the reference vehicle that provides a better match between the actual path traveled by the ego vehicle and the estimated path of the reference vehicle in relation to the traveled portion of the road segment. For example, in cases where the estimated path of the reference vehicle is directly coupled to the estimated size of a visual attribute associated with the reference vehicle (e.g., height, width, taillight size, or similar), a change in the estimated size provides a well-defined change to the distance. Petition 870250080995, dated 09 / 09 / 2025, page 47 / 151 38 / 56 estimated between the ego vehicle and the reference vehicle in each image. Consequently, in some respects, the onboard system can determine an updated value for the estimated measurement related to the size of the reference vehicle that provides the best match between the actual trajectory of the ego vehicle relative to a traveled portion of the road segment and the estimated trajectory of the reference vehicle relative to the same portion of the road segment. For example, in some respects, the best match between the actual trajectory of the ego vehicle and the estimated trajectory of the reference vehicle can be defined as an estimated trajectory that minimizes an error metric, such as a mean squared error, between the actual trajectory of the ego vehicle and the estimated trajectory of the reference vehicle.

[0057] For example, in Figure 4B, the bold line indicates the actual trajectory traveled by the ego vehicle, and the dashed lines indicate estimated trajectories of the reference vehicle that are based on different size estimates (e.g., Ho, ... Hi, ... Hn). In the illustrated example, the average value estimate (H = Hi) minimizes the error between the actual trajectory of the ego vehicle and the estimated trajectory of the reference vehicle, whereby the estimated size of the reference vehicle can be defined as Hi. Additionally or alternatively, in cases where a machine learning model is used to estimate the size of the reference vehicle and / or the distance between the ego vehicle and the reference vehicle, the correspondence between the actual trajectory traveled by the ego vehicle and the estimated trajectory of the reference vehicle can be used to compensate for systemic errors in the machine learning model. Petition 870250080995, dated 09 / 09 / 2025, page 48 / 151 39 / 56 machine. Furthermore, as shown, the updated size estimate can be used to update the surface geometry of the untraveled portion of the track segment ahead of the ego vehicle by updating the tracked trajectory of the reference vehicle (shown as a dashed line). In other words, the improved size estimate for a tracked reference vehicle can be used to improve historical and future distance and trajectory estimates for the tracked reference vehicle and any aggregate estimate of a surface geometry for a track segment that incorporates the estimated trajectory for the tracked reference vehicle. Additionally, in some cases, the accurate trajectory of the ego vehicle that is tracked along the traveled portion of the track segment can be extended within a threshold distance ahead of the vehicle (e.g., using structure from motion techniques).

[0058] Additionally or alternatively, in some respects, the estimated distance between the ego vehicle and the reference vehicle at the time a particular image of the reference vehicle is captured can be determined by relating the vehicle's position to a reference point or other visual attribute in the image that can be tracked relative to the image series (e.g., using computer vision). In such cases, when the ego vehicle reaches the tracked reference point or visual attribute, the onboard system can determine an accurate estimate of the distance between the ego vehicle and the reference vehicle at the time the image was captured. For example, when the onboard system observes the reference vehicle passing a signal. Petition 870250080995, dated 09 / 09 / 2025, page 49 / 151 40 / 56 stop sign (or any other visually traceable reference point or attribute) in the image series, the onboard system can track the stop sign until the ego vehicle reaches the stop sign and can determine an accurate estimate of the distance between the ego vehicle and the reference vehicle at the time the image of the reference vehicle passing the stop sign was captured (e.g., based on an estimate of the ego vehicle's position at the time the image of the reference vehicle passing the stop sign was captured and an estimate of the ego vehicle's position at the time the ego vehicle reached the stop sign).In this way, the accurate distance measurement can be used to update the size estimate for the reference vehicle, the estimated trajectory of the reference vehicle along an untraveled portion of the road segment ahead of the ego vehicle, and / or the estimated surface geometry for the untraveled portion of the road segment ahead of the ego vehicle.

[0059] As indicated above, Figures 4A to 4B are provided as an example. Other examples may differ from what is described in relation to Figures 4A and 4B.

[0060] Figure 5 is a flowchart of an example process 500 associated with estimating the road surface using vehicle detection. In some respects, one or more process blocks of Figure 5 are performed by one device (e.g., on-board system 120, on-board device 208, or similar). In some respects, one or more process blocks of Figure 5 are performed by another device or a group of devices, separate from or including the device, such as a network node (e.g., network node 150) and / or a remote device (e.g., Petition 870250080995, dated 09 / 09 / 2025, page 50 / 151 41 / 56 remote device 130). Additionally or alternatively, one or more process blocks of Figure 5 may be performed by device 300 or one or more components of device 300, such as processor 310, memory 315, input component 320, output component 325, communication component 330 and / or estimation component 335.

[0061] As shown in Figure 5, process 500 may include obtaining a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle (block 510). For example, the device may obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle, as described above.

[0062] As further shown in Figure 5, process 500 may include estimating, based on the image series, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle (block 520). For example, the device may estimate, based on the image series, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle, as described above.

[0063] As further shown in Figure 5, process 500 may include tracking a reference vehicle trajectory along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the image series (block 530). For example, the device may track a reference vehicle trajectory along the road segment ahead of the ego vehicle. Petition 870250080995, dated 09 / 09 / 2025, page 51 / 151 42 / 56 ego based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the image series, as described above.

[0064] As further shown in Figure 5, process 500 may include estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle (block 540). For example, the device may estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle, as described above.

[0065] Process 500 may include additional aspects, such as any single aspect or any combination of aspects described below and / or in conjunction with one or more of the other processes described elsewhere in the present invention.

[0066] In a first aspect, the estimated surface geometry associated with the track segment includes an alignment and a profile associated with the track segment.

[0067] In a second aspect, alone or in combination with the first aspect, the tracked trajectory of the reference vehicle is used to estimate the surface geometry associated with the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle with similar trajectories.

[0068] In a third aspect, alone or in combination with one or more of the first and second aspects, process 500 includes obtaining measurements that indicate an actual trajectory of the ego vehicle along a portion of the road segment traveled by the ego vehicle and updating the estimated size. Petition 870250080995, dated 09 / 09 / 2025, page 52 / 151 43 / 56 of the reference vehicle based on the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0069] In a fourth aspect, alone or in combination with one or more of the first through third aspects, process 500 includes updating the estimated surface geometry for an untraveled portion of the road segment based on the tracked trajectory of the reference vehicle along the untraveled portion of the road segment and the updated size of the reference vehicle.

[0070] In a fifth aspect, alone or in combination with one or more of the first to fourth aspects, updating the estimated size of the reference vehicle includes calculating a physical size of the reference vehicle that minimizes an error between the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0071] In a sixth aspect, alone or in combination with one or more of the first to fifth aspects, process 500 includes extending the actual trajectory of the ego vehicle to an untraveled portion of the road segment ahead of the ego vehicle using one or more movement technique structures.

[0072] In a seventh aspect, alone or in combination with one or more of the first to sixth aspects, process 500 involves associating the position of the reference vehicle relative to the image series with a reference point that is depicted in the image series and tracking the reference point relative to the image series, wherein the position of the reference vehicle relative to the ego vehicle is updated based on the ego vehicle reaching a Petition 870250080995, dated 09 / 09 / 2025, page 53 / 151 44 / 56 reference point position.

[0073] In an eighth aspect, alone or in combination with one or more of the first to seventh aspects, the reference vehicle is included among multiple reference vehicles associated with tracked trajectories along the road segment ahead of the ego vehicle, and the estimated surface geometry associated with the road segment is based on an aggregation of the tracked trajectories of the multiple reference vehicles.

[0074] In a ninth aspect, alone or in combination with one or more of the first through eighth aspects, estimating the position of the reference vehicle relative to the ego vehicle includes estimating, for each image in the image series, a distance between the ego vehicle and the reference vehicle based on one or more intrinsic parameters associated with a camera used to capture the image series, an estimated measurement related to the size of the reference vehicle, and a measurement related to the size of the reference vehicle that is measured in pixels based on the image.

[0075] In a tenth aspect, alone or in combination with one or more of the first to ninth aspects, the estimated distance between the ego vehicle and the reference vehicle is additionally based on a noise model.

[0076] In an eleventh aspect, alone or in combination with one or more of the first to tenth aspects, the estimated distance between the ego vehicle and the reference vehicle is additionally based on an estimated pose of the reference vehicle.

[0077] In a twelfth aspect, alone or in combination with one or more of the first to eleventh aspects. Petition 870250080995, dated 09 / 09 / 2025, p. 54 / 151 45 / 56 aspects, the 500 process includes tracking the trajectory of an object depicted in the image series based on the estimated surface geometry associated with the track segment ahead of the ego vehicle.

[0078] In a thirteenth aspect, alone or in combination with one or more of the first to twelfth aspects, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle are estimated using a machine learning model.

[0079] Although Figure 5 shows example blocks of process 500, in some respects, process 500 includes additional blocks, a smaller number of blocks, different blocks, or blocks arranged differently from those depicted in Figure 5. Additionally or alternatively, two or more of the blocks of process 500 can be performed in parallel.

[0080] The following is an overview of some aspects of this disclosure:

[0081] Aspect 1: A method for estimating road surface using vehicle detection, comprising: obtaining, by a device associated with an ego vehicle, a series of images depicting a reference vehicle moving along a road segment ahead of the ego vehicle; estimating, by the device based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; tracking, by the device, a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle. Petition 870250080995, dated 09 / 09 / 2025, page 55 / 151 46 / 56 reference vehicle in relation to the series of images; and estimate, using the device, a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0082] Aspect 2: The method of Aspect 1, in which the estimated surface geometry associated with the track segment includes an alignment and a profile associated with the track segment.

[0083] Aspect 3: The method of either Aspect 1 and 2, wherein the tracked trajectory of the reference vehicle is used to estimate the surface geometry associated with the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle with similar trajectories.

[0084] Aspect 4: The method of any of Aspects 1 to 3, further comprising: obtaining measurements that indicate an actual trajectory of the ego vehicle along a portion of the road segment traveled by the ego vehicle; and updating the estimated size of the reference vehicle based on the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0085] Aspect 5: The method of Aspect 4, further comprising: updating the estimated surface geometry for an untraveled portion of the road segment based on the tracked trajectory of the reference vehicle along the untraveled portion of the road segment and the updated size of the reference vehicle.

[0086] Aspect 6: The method of Aspect 4, in which the estimated size of the reference vehicle is updated. Petition 870250080995, dated 09 / 09 / 2025, page 56 / 151 47 / 56 includes: calculating a physical size of the reference vehicle that minimizes an error between the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0087] Aspect 7: The method of Aspect 4, additionally comprising: extending the actual trajectory of the ego vehicle to an untraveled portion of the road segment ahead of the ego vehicle using one or more movement technique structures.

[0088] Aspect 8: The method of Aspect 4, further comprising: associating the position of the reference vehicle relative to the image series with a reference point that is depicted in the image series; and tracking the reference point relative to the image series, where the position of the reference vehicle relative to the ego vehicle is updated based on the ego vehicle reaching a reference point position.

[0089] Aspect 9: The method of any of Aspects 1 to 8, wherein the reference vehicle is included among multiple reference vehicles associated with tracked trajectories along the road segment ahead of the ego vehicle, and wherein the estimated surface geometry associated with the road segment is based on an aggregation of the tracked trajectories of the multiple reference vehicles.

[0090] Aspect 10: The method of any of Aspects 1 to 9, wherein the estimation of the position of the reference vehicle relative to the ego vehicle includes: estimating, for each image in the image series, a distance between the ego vehicle and the reference vehicle based on one or more intrinsic parameters associated with a camera used. Petition 870250080995, dated 09 / 09 / 2025, page 57 / 151 48 / 56 to capture the series of images, an estimated measurement related to the size of the reference vehicle, and a measurement related to the size of the reference vehicle that is measured in pixels based on the image.

[0091] Aspect 11: The method of Aspect 10, in which the estimated distance between the ego vehicle and the reference vehicle is additionally based on a noise model.

[0092] Aspect 12: The method of Aspect 10, in which the estimated distance between the ego vehicle and the reference vehicle is additionally based on an estimated pose of the reference vehicle.

[0093] Aspect 13: The method of any of Aspects 1 to 12, additionally comprising: tracing a trajectory of an object depicted in the image series based on the estimated surface geometry associated with the road segment ahead of the ego vehicle.

[0094] Aspect 14: The method of any of Aspects 1 to 13, in which one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle are estimated using a machine learning model.

[0095] Aspect 15: A device for estimating road surface using vehicle detection, comprising: a memory; and one or more processors, coupled to the memory, configured to: obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; estimate, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; track Petition 870250080995, dated 09 / 09 / 2025, page 58 / 151 49 / 56 trace a reference vehicle trajectory along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the image series; and estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the traced trajectory of the reference vehicle.

[0096] Aspect 16: The Aspect 15 device, where the estimated surface geometry associated with the track segment includes an alignment and a profile associated with the track segment.

[0097] Aspect 17: The device of either Aspects 15 and 16, wherein the tracked trajectory of the reference vehicle is used to estimate the surface geometry associated with the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle with similar trajectories.

[0098] Aspect 18: The device of any of Aspects 15 to 17, in which one or more processors are additionally configured to: obtain measurements that indicate an actual trajectory of the ego vehicle along a portion of the road segment traveled by the ego vehicle; and update the estimated size of the reference vehicle based on the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0099] Aspect 19: The Aspect 18 device, in which one or more processors are additionally configured to: update the estimated surface geometry for an untraveled portion of the track segment. Petition 870250080995, dated 09 / 09 / 2025, page 59 / 151 50 / 56 based on the tracked trajectory of the reference vehicle along the untraveled portion of the road segment and the updated size of the reference vehicle.

[0100] Aspect 20: The Aspect 18 device, in which one or more processors, to update the estimated size of the reference vehicle, are configured to: calculate a physical size of the reference vehicle that minimizes an error between the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

[0101] Aspect 21: The Aspect 18 device, in which one or more processors are additionally configured to: extend the actual trajectory of the ego vehicle to an untraveled portion of the road segment ahead of the ego vehicle using one or more motion technique frameworks.

[0102] Aspect 22: The Aspect 18 device, wherein one or more processors are additionally configured to: associate the position of the reference vehicle relative to the image series with a reference point that is depicted in the image series; and track the reference point relative to the image series, wherein the position of the reference vehicle relative to the ego vehicle is updated based on the ego vehicle reaching a reference point position.

[0103] Aspect 23: The device of any of Aspects 15 to 22, in which the reference vehicle is included among multiple reference vehicles associated with tracked trajectories along the road segment ahead of the ego vehicle, and in which the estimated surface geometry associated with the road segment is based on an aggregation of the tracked trajectories of the multiple reference vehicles. Petition 870250080995, dated 09 / 09 / 2025, p. 60 / 151 51 / 56

[0104] Aspect 24: The device of any of Aspects 15 to 23, in which one or more processors, to estimate the position of the reference vehicle relative to the ego vehicle, are configured to: estimate, for each image in the image series, a distance between the ego vehicle and the reference vehicle based on one or more intrinsic parameters associated with a camera used to capture the image series, an estimated measurement related to the size of the reference vehicle, and a measurement related to the size of the reference vehicle that is measured in pixels based on the image.

[0105] Aspect 25: The Aspect 24 device, where the estimated distance between the ego vehicle and the reference vehicle is additionally based on a noise model.

[0106] Aspect 26: The Aspect 24 device, where the estimated distance between the ego vehicle and the reference vehicle is additionally based on an estimated pose of the reference vehicle.

[0107] Aspect 27: The device of any of Aspects 15 to 26, in which one or more processors are additionally configured to: track a trajectory of an object depicted in the image series based on the estimated surface geometry associated with the road segment ahead of the ego vehicle.

[0108] Aspect 28: The device of any of Aspects 15 to 27, wherein one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle are estimated using a machine learning model.

[0109] Aspect 29: A non-transient, computer-readable medium that stores a set of instructions, the Petition 870250080995, dated 09 / 09 / 2025, page 61 / 151 52 / 56 set of instructions comprising: one or more instructions that, when executed by one or more processors of a device, cause the device to: obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; estimate, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0110] Aspect 30: An apparatus comprising: means for obtaining a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; means for estimating, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; means for tracking a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and means for estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

[0111] Aspect 31: A system configured for Petition 870250080995, dated 09 / 09 / 2025, page 62 / 151 53 / 56 perform one or more operations mentioned in one or more of Aspects 1 to 30.

[0112] Aspect 32: An apparatus comprising means for carrying out one or more operations mentioned in one or more of Aspects 1 to 30.

[0113] Aspect 33: A non-transient, computer-readable medium that stores a set of instructions, the instruction set comprising one or more instructions which, when executed by a device, cause the device to perform one or more operations mentioned in one or more of Aspects 1 to 30.

[0114] Aspect 34: A computer program product comprising instructions or code for performing one or more operations mentioned in one or more of Aspects 1 to 30.

[0115] The aforementioned disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise forms disclosed. Modifications and variations may be made in light of the above disclosure or may be acquired from practice of the aspects.

[0116] As used in the present invention, the term component is intended to be broadly interpreted as hardware and / or a combination of hardware and software. The term software should be interpreted broadly to mean instructions, sets of instructions, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, execution threads, procedures and / or functions, among other examples, whether they are called... Petition 870250080995, dated 09 / 09 / 2025, page 63 / 151 54 / 56 software, firmware, middleware, microcode, hardware description language, or otherwise. As used in the present invention, a processor is implemented in hardware and / or in a combination of hardware and software. It will be evident that the systems and / or methods described in the present invention can be implemented in different forms of hardware and / or in a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and / or methods is not limiting. Thus, the operation and behavior of the systems and / or methods are described in the present invention without reference to specific software code, as those skilled in the art will understand that software and hardware can be designed to implement the systems and / or methods based, at least in part, on the description in the present invention.

[0117] As used in the present invention, satisfying a threshold may, depending on the context, refer to a value that is greater than the threshold, greater than or equal to the threshold, less than the threshold, less than or equal to the threshold, equal to the threshold, not equal to the threshold, or similar.

[0118] Although particular combinations of attributes are mentioned in the claims and / or disclosed in the descriptive report, these combinations are not intended to limit the disclosure of multiple aspects. Many of these attributes may be combined in ways not specifically mentioned in the claims and / or disclosed in the descriptive report. Disclosure of multiple aspects includes each dependent claim in combination with each other claim in the set of claims. Petition 870250080995, dated 09 / 09 / 2025, p. 64 / 151 55 / 56 As used in the present invention, an expression referring to at least one of a list of items refers to any combination of those items, including single members. For example, at least one of: a, b, or c is intended to cover a, b, c, a + b, a + c, b + c and a + b + c, as well as any combination with multiples of the same element (for example, a + a, a + a + a, a + a + b, a + a + c, a + b + b, a + c + c, b + b, b + b + b, b + b + c, c + c and c + c + c, or any other order of a, b and c).

[0119] No element, action, or instruction used in the present invention should be interpreted as critical or essential, except when explicitly described as such. Furthermore, as used in the present invention, the articles "a" and "an" are intended to include one or more items and may be used interchangeably with "one or more". Additionally, as used in the present invention, the article "the" or "a" is intended to include one or more items mentioned in conjunction with the article "the" or "a", and may be used interchangeably with "one or more" or "one or more". Furthermore, as used in the present invention, the terms "set" and "group" are intended to include one or more items and may be used interchangeably with "one or more". Where only one item is intended, the phrase "only one" or similar language is used.Furthermore, as used in the present invention, the terms "has," "have," "that has," or similar terms are intended to be non-limiting terms that do not restrict an element they modify (for example, an element that has A may also have B). Additionally, the phrase "based on" is intended to mean "based, at least in part, on," except where specifically indicated otherwise. Furthermore. Petition 870250080995, dated 09 / 09 / 2025, page 65 / 151 56 / 56 In addition, as used in the present invention, the term "or" is intended to be inclusive when used in a series and may be used interchangeably with "and / or," except where specifically indicated otherwise (for example, if used in combination with "either" or "the other" or "only one of them").

Claims

1. A method for estimating road surface using vehicle detection, characterized by comprising: obtaining, by a device associated with an ego vehicle, a series of images depicting a reference vehicle moving along a road segment ahead of the ego vehicle; estimating, by the device based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; tracking, by the device, a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and estimating, by the device, a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

2. A method according to claim 1, characterized in that the estimated surface geometry associated with the road segment includes an alignment and a profile associated with the road segment.

3. Method, according to claim 1, characterized in that the tracked trajectory of the reference vehicle is used to estimate the surface geometry associated with the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle with similar trajectories.

4. Method according to claim 1, Petition 870250080995, dated 09 / 09 / 2025, pp. 124 / 151 2 / 9 characterized by further comprising: obtaining measurements that indicate an actual trajectory of the ego vehicle along a portion of the road segment traveled by the ego vehicle; and updating the estimated size of the reference vehicle based on the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

5. A method according to claim 4, characterized by further comprising: updating the estimated surface geometry for an untraveled portion of the road segment based on the tracked trajectory of the reference vehicle along the untraveled portion of the road segment and the updated size of the reference vehicle.

6. Method, according to claim 4, characterized in that the update of the estimated size of the reference vehicle includes: calculating a physical size of the reference vehicle that minimizes an error between the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

7. Method, according to claim 4, characterized by further comprising: extending the actual trajectory of the ego vehicle to an untraveled portion of the road segment ahead of the ego vehicle using one or more motion technique structures.

8. Method according to claim 4, characterized by further comprising: associating the position of the reference vehicle in Petition 870250080995, dated 09 / 09 / 2025, page 125 / 151 3 / 9 in relation to the image series with a reference point that is depicted in the image series; and tracking the reference point in relation to the image series, wherein the position of the reference vehicle in relation to the ego vehicle is updated based on the ego vehicle reaching a reference point position.

9. A method according to claim 1, characterized in that the reference vehicle is included among multiple reference vehicles associated with tracked trajectories along the road segment ahead of the ego vehicle, and in that the estimated surface geometry associated with the road segment is based on an aggregation of the tracked trajectories of the multiple reference vehicles.

10. Method, according to claim 1, characterized in that the estimation of the position of the reference vehicle relative to the ego vehicle includes: estimating, for each image in the image series, a distance between the ego vehicle and the reference vehicle based on one or more intrinsic parameters associated with a camera used to capture the image series, an estimated measurement related to the size of the reference vehicle, and a measurement related to the size of the reference vehicle that is measured in pixels based on the image.

11. Method, according to claim 10, characterized in that the estimated distance between the ego vehicle and the reference vehicle is additionally based on a noise model.

12. Method, according to claim 10, characterized in that the estimated distance between the ego vehicle and the reference vehicle is based additionally on Petition 870250080995, dated 09 / 09 / 2025, pp. 126 / 151 4 / 9 an estimated pose of the reference vehicle.

13. Method, according to claim 1, characterized by further comprising: tracking a trajectory of an object depicted in the series of images based on the estimated surface geometry associated with the road segment ahead of the ego vehicle.

14. A method according to claim 1, characterized in that one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle are estimated using a machine learning model.

15. Device for estimating road surface using vehicle detection characterized by comprising: a memory; and one or more processors, coupled to the memory, configured to: obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; estimate, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle. Petition 870250080995, dated 09 / 09 / 2025, pp. 127 / 151 5 / 9 16. Device according to claim 15, characterized in that the estimated surface geometry associated with the track segment includes an alignment and a profile associated with the track segment.

17. Device, according to claim 15, characterized in that the tracked trajectory of the reference vehicle is used to estimate the surface geometry associated with the road segment ahead of the ego vehicle based on the reference vehicle and the ego vehicle with similar trajectories.

18. Device according to claim 15, characterized in that one or more processors are additionally configured to: obtain measurements that indicate an actual trajectory of the ego vehicle along a portion of the road segment traveled by the ego vehicle; and update the estimated size of the reference vehicle based on the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

19. Device according to claim 18, characterized in that one or more processors are additionally configured to: update the estimated surface geometry for an untraveled portion of the road segment based on the tracked trajectory of the reference vehicle along the untraveled portion of the road segment and the updated size of the reference vehicle.

20. Device, according to claim 18, characterized by one or more processors, to update Petition 870250080995, dated 09 / 09 / 2025, page 128 / 151 6 / 9 the estimated size of the reference vehicle, being configured to: calculate a physical size of the reference vehicle that minimizes an error between the actual trajectory of the ego vehicle and the tracked trajectory of the reference vehicle along the traveled portion of the road segment.

21. Device, according to claim 18, characterized in that one or more processors are additionally configured to: extend the actual trajectory of the ego vehicle to an untraveled portion of the road segment ahead of the ego vehicle using one or more motion technique structures.

22. Device according to claim 18, characterized in that one or more processors are additionally configured to: associate the position of the reference vehicle relative to the image series with a reference point that is depicted in the image series; and track the reference point relative to the image series, wherein the position of the reference vehicle relative to the ego vehicle is updated based on the ego vehicle reaching a reference point position.

23. Device according to claim 15, characterized in that the reference vehicle is included among multiple reference vehicles associated with tracked trajectories along the road segment ahead of the ego vehicle, and in that the estimated surface geometry associated with the road segment is based on an aggregation of the tracked trajectories of the multiple reference vehicles.

24. Device according to claim 15, Petition 870250080995, dated 09 / 09 / 2025, pp. 129 / 151 7 / 9 characterized in that one or more processors, for estimating the position of the reference vehicle relative to the ego vehicle, are configured to: estimate, for each image in the image series, a distance between the ego vehicle and the reference vehicle based on one or more intrinsic parameters associated with a camera used to capture the image series, an estimated measurement related to the size of the reference vehicle, and a measurement related to the size of the reference vehicle that is measured in pixels based on the image.

25. Device according to claim 24, characterized in that the estimated distance between the ego vehicle and the reference vehicle is additionally based on a noise model.

26. Device according to claim 24, characterized in that the estimated distance between the ego vehicle and the reference vehicle is additionally based on an estimated pose of the reference vehicle.

27. Device, according to claim 15, characterized in that one or more processors are additionally configured to: track the trajectory of an object depicted in the series of images based on the estimated surface geometry associated with the road segment ahead of the ego vehicle.

28. Device according to claim 15, characterized in that one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle are estimated using a machine learning model.

29. Non-transitory, computer-readable medium. Petition 870250080995, dated 09 / 09 / 2025, page.130 / 151 8 / 9 characterized by storing a set of instructions, wherein the instruction set comprises: one or more instructions that, when executed by one or more processors of a device, cause the device to: obtain a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; estimate, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; track a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and estimate a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

30. Apparatus characterized by comprising: means for obtaining a series of images depicting a reference vehicle moving along a road segment ahead of an ego vehicle; means for estimating, based on the series of images, one or more of the size of the reference vehicle or the position of the reference vehicle relative to the ego vehicle; means for tracking a trajectory of the reference vehicle along the road segment ahead of the ego vehicle based on one or more of the estimated size of the reference vehicle or the estimated position of the reference vehicle relative to the series of images; and means for estimating a surface geometry associated with the road segment ahead of the ego vehicle based on the tracked trajectory of the reference vehicle.

Images