Construction zone polygon representation learning via pseudo-labels and symbiotic post-processing

Abstract

Provided are methods for construction zone polygon representation learning via pseudo-labels and symbiotic post-processing, which can include generating pseudo-labeled data associated with construction zone polygons from samples of driving data containing construction zone object annotations. Some methods described also include training a machine learning model to predict construction zone segmentation masks based on the pseudo-labeled data and aligning the construction zone object annotations with the predicted construction zone segmentation masks. Systems and computer program products are also provided.

Description

Attorney Docket No. 46154-0571WO1 / I2023191Construction Zone Polygon Representation Learning via Pseudo-Labels and Symbiotic Post-ProcessingBACKGROUND

[0001] Hazards can occur within a road network, such as construction zones, potholes, debris, and the like. Map data may not reflect the road hazards. Further, the dynamic nature of construction zones can cause rapid changes to identified construction zones.BRIEF DESCRIPTION OF THE FIGURES

[0002] FIG. 1 is an example environment in which a vehicle including one or more components of an autonomous system can be implemented;

[0003] FIG. 2 is a diagram of one or more systems of a vehicle including an autonomous system;

[0004] FIG. 3 is a diagram of components of one or more devices and / or one or more systems of FIGS. 1 and 2;

[0005] FIG. 4A is a diagram of certain components of an autonomous system;

[0006] FIG. 4B is a diagram of an implementation of a neural network;

[0007] FIG. 4C and 4D are a diagram illustrating example operation of a CNN;

[0008] FIG. 5 shows a diagram of an implementation of construction zone polygon representation learning via pseudo-labels and symbiotic post-processing.

[0009] FIG. 6 shows a construction zone used to generate construction zone polygons or pseudo labels.

[0010] FIG. 7 shows training a deep learning based model for construction zone segmentation.

[0011] FIG. 8 shows the architecture of the deep learning based model for construction zone segmentation.

[0012] FIG. 9 shows the architecture of the deep learning based model for construction zone segmentation with symbiotic post-processing.

[0013] FIG. 10 shows the calculation of loss used in symbiotic post-processing.Attorney Docket No. 46154-0571WO1 / I2023191

[0014] FIG. 11 is a flowchart of a process for construction zone polygon representation learning via pseudo-labels and symbiotic post-processing.DETAILED DESCRIPTION

[0015] In the following description numerous specific details are set forth in order to provide a thorough understanding of the present disclosure for the purposes of explanation. It will be apparent, however, that the embodiments described by the present disclosure can be practiced without these specific details. In some instances, well-known structures and devices are illustrated in block diagram form in order to avoid unnecessarily obscuring aspects of the present disclosure.

[0016] Specific arrangements or orderings of schematic elements, such as those representing systems, devices, modules, instruction blocks, data elements, and / or the like are illustrated in the drawings for ease of description. However, it will be understood by those skilled in the art that the specific ordering or arrangement of the schematic elements in the drawings is not meant to imply that a particular order or sequence of processing, or separation of processes, is required unless explicitly described as such. Further, the inclusion of a schematic element in a drawing is not meant to imply that such element is required in all embodiments or that the features represented by such element may not be included in or combined with other elements in some embodiments unless explicitly described as such.

[0017] Further, where connecting elements such as solid or dashed lines or arrows are used in the drawings to illustrate a connection, relationship, or association between or among two or more other schematic elements, the absence of any such connecting elements is not meant to imply that no connection, relationship, or association can exist. In other words, some connections, relationships, or associations between elements are not illustrated in the drawings so as not to obscure the disclosure. In addition, for ease of illustration, a single connecting element can be used to represent multiple connections, relationships or associations between elements. For example, where a connecting element represents communication of signals, data, or instructions (e.g., “software instructions’ ), it should be understood by those skilled in the art that such element canAttorney Docket No. 46154-0571WO1 / I2023191represent one or multiple signal paths (e.g., a bus), as may be needed, to affect the communication.

[0018] Although the terms first, second, third, and / or the like are used to describe various elements, these elements should not be limited by these terms. The terms first, second, third, and / or the like are used only to distinguish one element from another. For example, a first contact could be termed a second contact and, similarly, a second contact could be termed a first contact without departing from the scope of the described embodiments. The first contact and the second contact are both contacts, but they are not the same contact.

[0019] The terminology used in the description of the various described embodiments herein is included for the purpose of describing particular embodiments only and is not intended to be limiting. As used in the description of the various described embodiments and the appended claims, the singular forms “a,” “an” and “the” are intended to include the plural forms as well and can be used interchangeably with “one or more” or “at least one,” unless the context clearly indicates otherwise. It will also be understood that the term “and / or” as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items. It will be further understood that the terms “includes,” “including,” “comprises,” and / or “comprising,” when used in this description specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0020] As used herein, the terms “communication” and “communicate” refer to at least one of the reception, receipt, transmission, transfer, provision, and / or the like of information (or information represented by, for example, data, signals, messages, instructions, commands, and / or the like). For one unit (e.g., a device, a system, a component of a device or system, combinations thereof, and / or the like) to be in communication with another unit means that the one unit is able to directly or indirectly receive information from and / or send (e.g., transmit) information to the other unit. This may refer to a direct or indirect connection that is wired and / or wireless in nature. Additionally, two units may be in communication with each other even though theAttorney Docket No. 46154-0571WO1 / I2023191information transmitted may be modified, processed, relayed, and / or routed between the first and second unit. For example, a first unit may be in communication with a second unit even though the first unit passively receives information and does not actively transmit information to the second unit. As another example, a first unit may be in communication with a second unit if at least one intermediary unit (e.g., a third unit located between the first unit and the second unit) processes information received from the first unit and transmits the processed information to the second unit. In some embodiments, a message may refer to a network packet (e.g., a data packet and / or the like) that includes data.

[0021] As used herein, the term “if is, optionally, construed to mean “when”, “upon”, “in response to determining," “in response to detecting," and / or the like, depending on the context. Similarly, the phrase “if it is determined” or “if [a stated condition or event] is detected” is, optionally, construed to mean “upon determining,” “in response to determining,” “upon detecting [the stated condition or event]," “in response to detecting [the stated condition or event],” and / or the like, depending on the context. Also, as used herein, the terms “has", “have”, “having”, or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based at least partially on” unless explicitly stated otherwise.

[0022] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the various described embodiments. However, it will be apparent to one of ordinary skill in the art that the various described embodiments can be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0023] General Overview

[0024] In some aspects and / or embodiments, systems, methods, and computer program products described herein include and / or implement construction zone polygon representation learning via pseudo-labels and symbiotic post-processing. A vehicle (such as an autonomous vehicle) avoids non-drivable segments corresponding to constructionAttorney Docket No. 46154-0571WO1 / I2023191zones by identifying construction zone polygons. A pseudo-label is a label attached to a particular datum or data that is determined by an automated process (i.e., not a human). Pseudo-labeled data is derived from driving data including bounding boxes associated with detected construction zone objects. A machine learning model is trained to predict construction zone polygons using pseudo-labeled data and human annotated data. Symbiotic post-processing is applied to the detected construction zone objects and predicted construction zone polygons. In some embodiments, the machine learning model is finetuned on a smaller dataset resulting from the symbiotic post-processing.

[0025] By virtue of the implementation of systems, methods, and computer program products described herein, techniques for construction zone polygon representation learning via pseudo-labels and symbiotic post-processing improves the identification of road hazards. For example, some of the advantages of these techniques include a more accurate identification of construction zones. The use of pseudo-labels increases the quantity of labeled data available to train a machine learning model to predict construction zones. The dual loss functions enable the optimization of multiple training objectives: parameters derived from 1) construction zone identification based on construction zone objects and 2) human identified ground truth construction zones, resulting in an accurate a robust model that predicts construction zones.

[0026] Referring now to FIG. 1, illustrated is example environment 100 in which vehicles that include autonomous systems, as well as vehicles that do not, are operated As illustrated, environment 100 includes vehicles 102a-102n, objects 104a-104n, routes 106a-106n, area 108, vehicle-to-infrastructure (V2I) device 110, network 112, remote autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118. Vehicles 102a-102n, vehicle-to-infrastructure (V2I) device 110, network 112, autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118 interconnect (e.g., establish a connection to communicate and / or the like) via wired connections, wireless connections, or a combination of wired or wireless connections. In some embodiments, objects 104a-104n interconnect with at least one of vehicles 102a- 102n, vehicle-to-infrastructure (V2I) device 110, network 112, autonomous vehicle (AV) system 114, fleet management system 116, and V2I system 118 via wired connections, wireless connections, or a combination of wired or wireless connections.Attorney Docket No. 46154-0571WO1 / I2023191

[0027] Vehicles 102a-102n (referred to individually as vehicle 102 and collectively as vehicles 102) include at least one device configured to transport goods and / or people. In some embodiments, vehicles 102 are configured to be in communication with V2I device 110, remote AV system 114, fleet management system 116, and / or V2I system 118 via network 112. In some embodiments, vehicles 102 include cars, buses, trucks, trains, and / or the like. In some embodiments, vehicles 102 are the same as, or similar to, vehicles 200, described herein (see FIG. 2). In some embodiments, a vehicle 200 of a set of vehicles 200 is associated with an autonomous fleet manager. In some embodiments, vehicles 102 travel along respective routes 106a-106n (referred to individually as route 106 and collectively as routes 106), as described herein. In some embodiments, one or more vehicles 102 include an autonomous system (e.g., an autonomous system that is the same as or similar to autonomous system 202).

[0028] Objects 104a-104n (referred to individually as object 104 and collectively as objects 104) include, for example, at least one vehicle, at least one pedestrian, at least one cyclist, at least one structure (e.g., a building, a sign, a fire hydrant, etc.), and / or the like. Each object 104 is stationary (e.g., located at a fixed location for a period of time) or mobile (e.g., having a velocity and associated with at least one trajectory). In some embodiments, objects 104 are associated with corresponding locations in area 108.

[0029] Routes 106a-106n (referred to individually as route 106 and collectively as routes 106) are each associated with (e.g., prescribe) a sequence of actions (also known as a trajectory) connecting states along which an AV can navigate. Each route 106 starts at an initial state (e.g., a state that corresponds to a first spatiotemporal location, velocity, and / or the like) and ends at a final goal state (e.g., a state that corresponds to a second spatiotemporal location that is different from the first spatiotemporal location) or goal region (e.g. a subspace of acceptable states (e.g., terminal states)). In some embodiments, the first state includes a location at which an individual or individuals are to be picked-up by the AV and the second state or region includes a location or locations at which the individual or individuals picked-up by the AV are to be dropped-off. In some embodiments, routes 106 include a plurality of acceptable state sequences (e.g., a plurality of spatiotemporal location sequences), the plurality of state sequences associated with (e.g., defining) a plurality of trajectories. In an example, routes 106Attorney Docket No. 46154-0571WO1 / I2023191include only high level actions or imprecise state locations, such as a series of connected roads dictating turning directions at roadway intersections. Additionally, or alternatively, routes 106 may include more precise actions or states such as, for example, specific target lanes or precise locations within the lane areas and targeted speed at those positions. In an example, routes 106 include a plurality of precise state sequences along the at least one high level action sequence with a limited lookahead horizon to reach intermediate goals, where the combination of successive iterations of limited horizon state sequences cumulatively correspond to a plurality of trajectories that collectively form the high level route to terminate at the final goal state or region.

[0030] Area 108 includes a physical area (e.g., a geographic region) within which vehicles 102 can navigate. In an example, area 108 includes at least one state (e.g., a country, a province, an individual state of a plurality of states included in a country, etc.), at least one portion of a state, at least one city, at least one portion of a city, etc. In some embodiments, area 108 includes at least one named thoroughfare (referred to herein as a “road”) such as a highway, an interstate highway, a parkway, a city street, etc. Additionally, or alternatively, in some examples area 108 includes at least one unnamed road such as a driveway, a section of a parking lot, a section of a vacant and / or undeveloped lot, a dirt path, etc. In some embodiments, a road includes at least one lane (e.g., a portion of the road that can be traversed by vehicles 102). In an example, a road includes at least one lane associated with (e.g., identified based on) at least one lane marking.

[0031] Vehicle-to-lnfrastructure (V2I) device 110 (sometimes referred to as a Vehicle-to-lnfrastructure or Vehicle-to-Everything (V2X) device) includes at least one device configured to be in communication with vehicles 102 and / or V2I infrastructure system 118. In some embodiments, V2I device 110 is configured to be in communication with vehicles 102, remote AV system 114, fleet management system 116, and / or V2I system 118 via network 112. In some embodiments, V2I device 110 includes a radio frequency identification (RFID) device, signage, cameras (e.g., two-dimensional (2D) and / or three-dimensional (3D) cameras), lane markers, streetlights, parking meters, etc. In some embodiments, V2I device 110 is configured to communicate directly with vehicles 102. Additionally, or alternatively, in some embodiments V2I device 110 is configured toAttorney Docket No. 46154-0571WO1 / I2023191communicate with vehicles 102, remote AV system 114, and / or fleet management system 116 via V2I system 118. In some embodiments, V2I device 110 is configured to communicate with V2I system 118 via network 112.

[0032] Network 112 includes one or more wired and / or wireless networks. In an example, network 112 includes a cellular network (e.g., a long term evolution (LTE) network, a third generation (3G) network, a fourth generation (4G) network, a fifth generation (5G) network, a code division multiple access (CDMA) network, etc.), 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., the public switched telephone network (PSTN), a private network, an ad hoc network, an intranet, the Internet, a fiber optic-based network, a cloud computing network, etc., a combination of some or all of these networks, and / or the like.

[0033] Remote AV system 114 includes at least one device configured to be in communication with vehicles 102, V2I device 110, network 112, fleet management system 116, and / or V2I system 118 via network 112. In an example, remote AV system 114 includes a server, a group of servers, and / or other like devices. In some embodiments, remote AV system 114 is co-located with the fleet management system 116. In some embodiments, remote AV system 114 is involved in the installation of some or all of the components of a vehicle, including an autonomous system, an autonomous vehicle compute, software implemented by an autonomous vehicle compute, and / or the like. In some embodiments, remote AV system 114 maintains (e.g., updates and / or replaces) such components and / or software during the lifetime of the vehicle.

[0034] Fleet management system 116 includes at least one device configured to be in communication with vehicles 102, V2I device 110, remote AV system 114, and / or V2I infrastructure system 118. In an example, fleet management system 116 includes a server, a group of servers, and / or other like devices. In some embodiments, fleet management system 116 is associated with a ridesharing company (e.g., an organization that controls operation of multiple vehicles (e.g., vehicles that include autonomous systems and / or vehicles that do not include autonomous systems) and / or the like).

[0035] In some embodiments, V2I system 118 includes at least one device configured to be in communication with vehicles 102, V2I device 110, remote AV systemAttorney Docket No. 46154-0571WO1 / I2023191114, and / or fleet management system 116 via network 112. In some examples, V2I system 118 is configured to be in communication with V2I device 110 via a connection different from network 112. In some embodiments, V2I system 118 includes a server, a group of servers, and / or other like devices. In some embodiments, V2I system 118 is associated with a municipality or a private institution (e.g., a private institution that maintains V2I device 110 and / or the like).

[0036] The number and arrangement of elements illustrated in FIG. 1 are provided as an example. There can be additional elements, fewer elements, different elements, and / or differently arranged elements, than those illustrated in FIG. 1. Additionally, or alternatively, at least one element of environment 100 can perform one or more functions described as being performed by at least one different element of FIG. 1. Additionally, or alternatively, at least one set of elements of environment 100 can perform one or more functions described as being performed by at least one different set of elements of environment 100.

[0037] Referring now to FIG. 2, vehicle 200 (which may be the same as, or similar to vehicles 102 of FIG. 1) includes or is associated with autonomous system 202, powertrain control system 204, steering control system 206, and brake system 208. In some embodiments, vehicle 200 is the same as or similar to vehicle 102 (see FIG. 1 ). In some embodiments, autonomous system 202 is configured to confer vehicle 200 autonomous driving capability (e.g., implement at least one driving automation or maneuver-based function, feature, device, and / or the like that enable vehicle 200 to be partially or fully operated without human intervention including, without limitation, fully autonomous vehicles (e.g., vehicles that forego reliance on human intervention such as Level 5 ADS-operated vehicles), highly autonomous vehicles (e.g., vehicles that forego reliance on human intervention in certain situations such as Level 4 ADS-operated vehicles), conditional autonomous vehicles (e.g., vehicles that forego reliance on human intervention in limited situations such as Level 3 ADS-operated vehicles) and / or the like. In one embodiment, autonomous system 202 includes operational or tactical functionality required to operate vehicle 200 in on-road traffic and perform part or all of Dynamic Driving Task (DDT) on a sustained basis. In another embodiment, autonomous system 202 includes an Advanced Driver Assistance System (ADAS) that includes driver supportAttorney Docket No. 46154-0571WO1 / I2023191features. Autonomous system 202 supports various levels of driving automation, ranging from no driving automation (e.g., Level 0) to full driving automation (e.g., Level 5). For a detailed description of fully autonomous vehicles and highly autonomous vehicles, reference may be made to SAE International's standard J3016: Taxonomy and Definitions for Terms Related to On-Road Motor Vehicle Automated Driving Systems, which is incorporated by reference in its entirety. In some embodiments, vehicle 200 is associated with an autonomous fleet manager and / or a ridesharing company.

[0038] Autonomous system 202 includes a sensor suite that includes one or more devices such as cameras 202a, LIDAR sensors 202b, radar sensors 202c, and microphones 202d. In some embodiments, autonomous system 202 can include more or fewer devices and / or different devices (e.g., ultrasonic sensors, inertial sensors, GPS receivers (discussed below), odometry sensors that generate data associated with an indication of a distance that vehicle 200 has traveled, and / or the like). In some embodiments, autonomous system 202 uses the one or more devices included in autonomous system 202 to generate data associated with environment 100, described herein. The data generated by the one or more devices of autonomous system 202 can be used by one or more systems described herein to observe the environment (e.g., environment 100) in which vehicle 200 is located. In some embodiments, autonomous system 202 includes communication device 202e, autonomous vehicle compute 202f, drive-by-wire (DBW) system 202h, and safety controller 202g.

[0039] Cameras 202a include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and / or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Cameras 202a include at least one camera (e.g., a digital camera using a light sensor such as a Charge-Coupled Device (CCD), a thermal camera, an Infrared (IR) camera, an event camera, and / or the like) to capture images including physical objects (e.g., cars, buses, curbs, people, and / or the like). In some embodiments, camera 202a generates camera data as output. In some examples, camera 202a generates camera data that includes image data associated with an image. In this example, the image data may specify at least one parameter (e.g., image characteristics such as exposure, brightness, etc., an image timestamp, and / or the like) corresponding to the image. In suchAttorney Docket No. 46154-0571WO1 / I2023191an example, the image may be in a format (e.g., RAW, JPEG, PNG, and / or the like). In some embodiments, camera 202a includes a plurality of independent cameras configured on (e.g., positioned on) a vehicle to capture images for the purpose of stereopsis (stereo vision). In some examples, camera 202a includes a plurality of cameras that generate image data and transmit the image data to autonomous vehicle compute 202f and / or a fleet management system (e.g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1). In such an example, autonomous vehicle compute 202f determines depth to one or more objects in a field of view of at least two cameras of the plurality of cameras based on the image data from the at least two cameras. In some embodiments, cameras 202a is configured to capture images of objects within a distance from cameras 202a (e.g., up to 100 meters, up to a kilometer, and / or the like). Accordingly, cameras 202a include features such as sensors and lenses that are optimized for perceiving objects that are at one or more distances from cameras 202a.

[0040] In an embodiment, camera 202a includes at least one camera configured to capture one or more images associated with one or more traffic lights, street signs and / or other physical objects that provide visual navigation information. In some embodiments, camera 202a generates traffic light data associated with one or more images. In some examples, camera 202a generates TLD (Traffic Light Detection) data associated with one or more images that include a format (e.g., RAW, JPEG, PNG, and / or the like). In some embodiments, camera 202a that generates TLD data differs from other systems described herein incorporating cameras in that camera 202a can include one or more cameras with a wide field of view (e.g., a wide-angle lens, a fish-eye lens, a lens having a viewing angle of approximately 120 degrees or more, and / or the like) to generate images about as many physical objects as possible.

[0041] Light Detection and Ranging (LiDAR) sensors 202b include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and / or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). LiDAR sensors 202b include a system configured to transmit light from a light emitter (e.g., a laser transmitter). Light emitted by LiDAR sensors 202b include light (e.g., infrared light and / or the like) that is outside of the visible spectrum.Attorney Docket No. 46154-0571WO1 / I2023191In some embodiments, during operation, light emitted by LIDAR sensors 202b encounters a physical object (e.g., a vehicle) and is reflected back to LiDAR sensors 202b. In some embodiments, the light emitted by LiDAR sensors 202b does not penetrate the physical objects that the light encounters. LiDAR sensors 202b also include at least one light detector which detects the light that was emitted from the light emitter after the light encounters a physical object. In some embodiments, at least one data processing system associated with LiDAR sensors 202b generates an image (e.g., a point cloud, a combined point cloud, and / or the like) representing the objects included in a field of view of LiDAR sensors 202b. In some examples, the at least one data processing system associated with LiDAR sensor 202b generates an image that represents the boundaries of a physical object, the surfaces (e.g., the topology of the surfaces) of the physical object, and / or the like. In such an example, the image is used to determine the boundaries of physical objects in the field of view of LiDAR sensors 202b.

[0042] Radio Detection and Ranging (radar) sensors 202c include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f, and / or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Radar sensors 202c include a system configured to transmit radio waves (either pulsed or continuously). The radio waves transmitted by radar sensors 202c include radio waves that are within a predetermined spectrum. In some embodiments, during operation, radio waves transmitted by radar sensors 202c encounter a physical object and are reflected back to radar sensors 202c. In some embodiments, the radio waves transmitted by radar sensors 202c are not reflected by some objects. In some embodiments, at least one data processing system associated with radar sensors 202c generates signals representing the objects included in a field of view of radar sensors 202c. For example, the at least one data processing system associated with radar sensor 202c generates an image that represents the boundaries of a physical object, the surfaces (e.g., the topology of the surfaces) of the physical object, and / or the like. In some examples, the image is used to determine the boundaries of physical objects in the field of view of radar sensors 202c.

[0043] Microphones 202d includes at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f,Attorney Docket No. 46154-0571WO1 / I2023191and / or safety controller 202g via a bus (e.g., a bus that is the same as or similar to bus 302 of FIG. 3). Microphones 202d include one or more microphones (e.g., array microphones, external microphones, and / or the like) that capture audio signals and generate data associated with (e.g., representing) the audio signals. In some examples, microphones 202d include transducer devices and / or like devices. In some embodiments, one or more systems described herein can receive the data generated by microphones 202d and determine a position of an object relative to vehicle 200 (e.g., a distance and / or the like) based on the audio signals associated with the data.

[0044] Communication device 202e includes at least one device configured to be in communication with cameras 202a, LIDAR sensors 202b, radar sensors 202c, microphones 202d, autonomous vehicle compute 202f, safety controller 202g, and / or DBW (Drive-By-Wire) system 202h. For example, communication device 202e may include a device that is the same as or similar to communication interface 314 of FIG. 3. In some embodiments, communication device 202e includes a vehicle-to-vehicle (V2V) communication device (e.g., a device that enables wireless communication of data between vehicles).

[0045] Autonomous vehicle compute 202f include at least one device configured to be in communication with cameras 202a, LIDAR sensors 202b, radar sensors 202c, microphones 202d, communication device 202e, safety controller 202g, and / or DBW system 202h. In some examples, autonomous vehicle compute 202f includes a device such as a client device, a mobile device (e.g., a cellular telephone, a tablet, and / or the like), a server (e.g., a computing device including one or more central processing units, graphical processing units, and / or the like), and / or the like. In some embodiments, autonomous vehicle compute 202f is the same as or similar to autonomous vehicle compute 400, described herein. Additionally, or alternatively, in some embodiments autonomous vehicle compute 202f is configured to be in communication with an autonomous vehicle system (e.g., an autonomous vehicle system that is the same as or similar to remote AV system 114 of FIG. 1), a fleet management system (e.g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1), a V2I device (e.g., a V2I device that is the same as or similar to V2I device 110Attorney Docket No. 46154-0571WO1 / I2023191of FIG. 1), and / or a V2I system (e.g., a V2I system that is the same as or similar to V2I system 118 of FIG. 1).

[0046] Safety controller 202g includes at least one device configured to be in communication with cameras 202a, LiDAR sensors 202b, radar sensors 202c, microphones 202d, communication device 202e, autonomous vehicle computer 202f, and / or DBW system 202h. In some examples, safety controller 202g includes one or more controllers (electrical controllers, electromechanical controllers, and / or the like) that are configured to generate and / or transmit control signals to operate one or more devices of vehicle 200 (e.g., powertrain control system 204, steering control system 206, brake system 208, and / or the like). In some embodiments, safety controller 202g is configured to generate control signals that take precedence over (e.g., overrides) control signals generated and / or transmitted by autonomous vehicle compute 202f.

[0047] DBW system 202h includes at least one device configured to be in communication with communication device 202e and / or autonomous vehicle compute 202f. In some examples, DBW system 202h includes one or more controllers (e.g., electrical controllers, electromechanical controllers, and / or the like) that are configured to generate and / or transmit control signals to operate one or more devices of vehicle 200 (e.g., powertrain control system 204, steering control system 206, brake system 208, and / or the like). Additionally, or alternatively, the one or more controllers of DBW system 202h are configured to generate and / or transmit control signals to operate at least one different device (e.g., a turn signal, headlights, door locks, windshield wipers, and / or the like) of vehicle 200.

[0048] Powertrain control system 204 includes at least one device configured to be in communication with DBW system 202h. In some examples, powertrain control system 204 includes at least one controller, actuator, and / or the like. In some embodiments, powertrain control system 204 receives control signals from DBW system 202h and powertrain control system 204 causes vehicle 200 to make longitudinal vehicle motion, such as start moving forward, stop moving forward, start moving backward, stop moving backward, accelerate in a direction, decelerate in a direction or to make lateral vehicle motion such as performing a left turn, performing a right turn, and / or the like. In an example, powertrain control system 204 causes the energy (e.g., fuel, electricity, and / orAttorney Docket No. 46154-0571WO1 / I2023191the like) provided to a motor of the vehicle to increase, remain the same, or decrease, thereby causing at least one wheel of vehicle 200 to rotate or not rotate.

[0049] Steering control system 206 includes at least one device configured to rotate one or more wheels of vehicle 200. In some examples, steering control system 206 includes at least one controller, actuator, and / or the like. In some embodiments, steering control system 206 causes the front two wheels and / or the rear two wheels of vehicle 200 to rotate to the left or right to cause vehicle 200 to turn to the left or right. In other words, steering control system 206 causes activities necessary for the regulation of the y-axis component of vehicle motion.

[0050] Brake system 208 includes at least one device configured to actuate one or more brakes to cause vehicle 200 to reduce speed and / or remain stationary. In some examples, brake system 208 includes at least one controller and / or actuator that is configured to cause one or more calipers associated with one or more wheels of vehicle 200 to close on a corresponding rotor of vehicle 200. Additionally, or alternatively, in some examples brake system 208 includes an automatic emergency braking (AEB) system, a regenerative braking system, and / or the like.

[0051] In some embodiments, vehicle 200 includes at least one platform sensor (not explicitly illustrated) that measures or infers properties of a state or a condition of vehicle 200. In some examples, vehicle 200 includes platform sensors such as a global positioning system (GPS) receiver, an inertial measurement unit (IMU), a wheel speed sensor, a wheel brake pressure sensor, a wheel torque sensor, an engine torque sensor, a steering angle sensor, and / or the like. Although brake system 208 is illustrated to be located in the near side of vehicle 200 in FIG. 2, brake system 208 may be located anywhere in vehicle 200.

[0052] Referring now to FIG. 3, illustrated is a schematic diagram of a device 300. As illustrated, device 300 includes processor 304, memory 306, storage component 308, input interface 310, output interface 312, communication interface 314, and bus 302. In some embodiments, device 300 corresponds to at least one device of vehicles 102 (e.g., at least one device of a system of vehicles 102) and / or one or more devices of network 112 (e.g., one or more devices of a system of network 112). In some embodiments, one or more devices of vehicles 102 (e.g., one or more devices of a system of vehicles 102)Attorney Docket No. 46154-0571WO1 / I2023191and / or one or more devices of network 112 (e.g., one or more devices of a system of network 112) include at least one device 300 and / or at least one component of device 300. As shown in FIG. 3, device 300 includes bus 302, processor 304, memory 306, storage component 308, input interface 310, output interface 312, and communication interface 314.

[0053] Bus 302 includes a component that permits communication among the components of device 300. In some cases, processor 304 includes a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU), an accelerated processing unit (APU), and / or the like), a microphone, a digital signal processor (DSP), and / or any processing component (e.g., a field-programmable gate array (FPGA), an application specific integrated circuit (ASIC), and / or the like) that can be programmed to perform at least one function. Memory 306 includes random access memory (RAM), read¬ only memory (ROM), and / or another type of dynamic and / or static storage device (e.g., flash memory, magnetic memory, optical memory, and / or the like) that stores data and / or instructions for use by processor 304.

[0054] Storage component 308 stores data and / or software related to the operation and use of device 300. In some examples, storage component 308 includes a hard disk (e.g., a magnetic disk, an optical disk, a magneto-optic disk, a solid state disk, and / or the like), a compact disc (CD), a digital versatile disc (DVD), a floppy disk, a cartridge, a magnetic tape, a CD-ROM, RAM, PROM, EPROM, FLASH-EPROM, NV-RAM, and / or another type of computer readable medium, along with a corresponding drive.

[0055] Input interface 310 includes a component that permits device 300 to receive information, such as via user input (e.g., a touchscreen display, a keyboard, a keypad, a mouse, a button, a switch, a microphone, a camera, and / or the like). Additionally or alternatively, in some embodiments input interface 310 includes a sensor that senses information (e.g., a global positioning system (GPS) receiver, an accelerometer, a gyroscope, an actuator, and / or the like). Output interface 312 includes a component that provides output information from device 300 (e.g., a display, a speaker, one or more light¬ emitting diodes (LEDs), and / or the like).

[0056] In some embodiments, communication interface 314 includes a transceiver¬ like component (e.g., a transceiver, a separate receiver and transmitter, and / or the like)Attorney Docket No. 46154-0571WO1 / I2023191that permits device 300 to communicate with other devices via a wired connection, a wireless connection, or a combination of wired and wireless connections. In some examples, communication interface 314 permits device 300 to receive information from another device and / or provide information to another device. In some examples, communication interface 314 includes an Ethernet interface, an optical interface, a coaxial interface, an infrared interface, a radio frequency (RF) interface, a universal serial bus (USB) interface, a Wi-Fi® interface, a cellular network interface, and / or the like.

[0057] In some embodiments, device 300 performs one or more processes described herein. Device 300 performs these processes based on processor 304 executing software instructions stored by a computer-readable medium, such as memory 305 and / or storage component 308. A computer-readable medium (e.g., a non-transitory computer readable medium) is defined herein as a non-transitory memory device. A non-transitory memory device includes memory space located inside a single physical storage device or memory space spread across multiple physical storage devices.

[0058] In some embodiments, software instructions are read into memory 306 and / or storage component 308 from another computer-readable medium or from another device via communication interface 314. When executed, software instructions stored in memory 306 and / or storage component 308 cause processor 304 to perform one or more processes described herein. Additionally or alternatively, hardwired circuitry is used in place of or in combination with software instructions to perform one or more processes described herein. Thus, embodiments described herein are not limited to any specific combination of hardware circuitry and software unless explicitly stated otherwise.

[0059] Memory 306 and / or storage component 308 includes data storage or at least one data structure (e.g., a database and / or the like). Device 300 is capable of receiving information from, storing information in, communicating information to, or searching information stored in the data storage or the at least one data structure in memory 306 or storage component 308. In some examples, the information includes network data, input data, output data, or any combination thereof.

[0060] In some embodiments, device 300 is configured to execute software instructions that are either stored in memory 306 and / or in the memory of another device (e.g., another device that is the same as or similar to device 300). As used herein, theAttorney Docket No. 46154-0571WO1 / I2023191term “module” refers to at least one instruction stored in memory 306 and / or in the memory of another device that, when executed by processor 304 and / or by a processor of another device (e.g., another device that is the same as or similar to device 300) cause device 300 (e.g., at least one component of device 300) to perform one or more processes described herein. In some embodiments, a module is implemented in software, firmware, hardware, and / or the like.

[0061] The number and arrangement of components illustrated in FIG. 3 are provided as an example. In some embodiments, device 300 can include additional components, fewer components, different components, or differently arranged components than those illustrated in FIG. 3. Additionally or alternatively, a set of components (e.g., one or more components) of device 300 can perform one or more functions described as being performed by another component or another set of components of device 300.

[0062] Referring now to FIG. 4A, illustrated is an example block diagram of an autonomous vehicle compute 400 (sometimes referred to as an “AV stack”). As illustrated, autonomous vehicle compute 400 includes perception system 402 (sometimes referred to as a perception module), planning system 404 (sometimes referred to as a planning module), localization system 406 (sometimes referred to as a localization module), control system 408 (sometimes referred to as a control module), and database 410. In some embodiments, perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included and / or implemented in an autonomous navigation system of a vehicle (e.g., autonomous vehicle compute 202f of vehicle 200). Additionally, or alternatively, in some embodiments perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included in one or more standalone systems (e.g., one or more systems that are the same as or similar to autonomous vehicle compute 400 and / or the like). In some examples, perception system 402, planning system 404, localization system 406, control system 408, and database 410 are included in one or more standalone systems that are located in a vehicle and / or at least one remote system as described herein. In some embodiments, any and / or all of the systems included in autonomous vehicle compute 400 are implemented in software (e.g., in software instructions stored in memory), computerAttorney Docket No. 46154-0571WO1 / I2023191hardware (e.g., by microprocessors, microcontrollers, application-specific integrated circuits (ASICs), Field Programmable Gate Arrays (FPGAs), and / or the like), or combinations of computer software and computer hardware. It will also be understood that, in some embodiments, autonomous vehicle compute 400 is configured to be in communication with a remote system (e.g., an autonomous vehicle system that is the same as or similar to remote AV system 114, a fleet management system 116 that is the same as or similar to fleet management system 116, a V2I system that is the same as or similar to V2I system 118, and / or the like).

[0063] In some embodiments, perception system 402 receives data associated with at least one physical object (e.g., data that is used by perception system 402 to detect the at least one physical object) in an environment and classifies the at least one physical object. In some examples, perception system 402 receives image data captured by at least one camera (e.g., cameras 202a), the image associated with (e.g., representing) one or more physical objects within a field of view of the at least one camera. In such an example, perception system 402 classifies at least one physical object based on one or more groupings of physical objects (e.g., bicycles, vehicles, traffic signs, pedestrians, and / or the like). In some embodiments, perception system 402 transmits data associated with the classification of the physical objects to planning system 404 based on perception system 402 classifying the physical objects.

[0064] In some embodiments, planning system 404 receives data associated with a destination and generates data associated with at least one route (e.g., routes 106) along which a vehicle (e.g., vehicles 102) can travel along toward a destination. In some embodiments, planning system 404 periodically or continuously receives data from perception system 402 (e.g., data associated with the classification of physical objects, described above) and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by perception system 402. In other words, planning system 404 may perform tactical function-related tasks that are required to operate vehicle 102 in on-road traffic. Tactical efforts involve maneuvering the vehicle in traffic during a trip, including but not limited to deciding whether and when to overtake another vehicle, change lanes, or selecting an appropriate speed, acceleration, deacceleration, etc. In some embodiments, planning system 404Attorney Docket No. 46154-0571WO1 / I2023191receives data associated with an updated position of a vehicle (e.g., vehicles 102) from localization system 406 and planning system 404 updates the at least one trajectory or generates at least one different trajectory based on the data generated by localization system 406.

[0065] In some embodiments, localization system 406 receives data associated with (e.g., representing) a location of a vehicle (e.g., vehicles 102) in an area. In some examples, localization system 406 receives LiDAR data associated with at least one point cloud generated by at least one LiDAR sensor (e.g., LiDAR sensors 202b). In certain examples, localization system 406 receives data associated with at least one point cloud from multiple LiDAR sensors and localization system 406 generates a combined point cloud based on each of the point clouds. In these examples, localization system 406 compares the at least one point cloud or the combined point cloud to two-dimensional (2D) and / or a three-dimensional (3D) map of the area stored in database 410. Localization system 406 then determines the position of the vehicle in the area based on localization system 406 comparing the at least one point cloud or the combined point cloud to the map. In some embodiments, the map includes a combined point cloud of the area generated prior to navigation of the vehicle. In some embodiments, maps include, without limitation, high-precision maps of the roadway geometric properties, maps describing road network connectivity properties, maps describing roadway physical properties (such as traffic speed, traffic volume, the number of vehicular and cyclist traffic lanes, lane width, lane traffic directions, or lane marker types and locations, or combinations thereof), and maps describing the spatial locations of road features such as crosswalks, traffic signs or other travel signals of various types. In some embodiments, the map is generated in real-time based on the data received by the perception system.

[0066] In another example, localization system 406 receives Global Navigation Satellite System (GNSS) data generated by a global positioning system (GPS) receiver. In some examples, localization system 406 receives GNSS data associated with the location of the vehicle in the area and localization system 406 determines a latitude and longitude of the vehicle in the area. In such an example, localization system 406 determines the position of the vehicle in the area based on the latitude and longitude of the vehicle. In some embodiments, localization system 406 generates data associatedAttorney Docket No. 46154-0571WO1 / I2023191with the position of the vehicle. In some examples, localization system 406 generates data associated with the position of the vehicle based on localization system 406 determining the position of the vehicle. In such an example, the data associated with the position of the vehicle includes data associated with one or more semantic properties corresponding to the position of the vehicle.

[0067] In some embodiments, control system 408 receives data associated with at least one trajectory from planning system 404 and control system 408 controls operation of the vehicle. In some examples, control system 408 receives data associated with at least one trajectory from planning system 404 and control system 408 controls operation of the vehicle by generating and transmitting control signals to cause a powertrain control system (e.g., DBW system 202h, powertrain control system 204, and / or the like), a steering control system (e.g., steering control system 206), and / or a brake system (e.g., brake system 208) to operate. For example, control system 408 is configured to perform operational functions such as a lateral vehicle motion control or a longitudinal vehicle motion control. The lateral vehicle motion control causes activities necessary for the regulation of the y-axis component of vehicle motion. The longitudinal vehicle motion control causes activities necessary for the regulation of the x-axis component of vehicle motion. In an example, where a trajectory includes a left turn, control system 408 transmits a control signal to cause steering control system 206 to adjust a steering angle of vehicle 200, thereby causing vehicle 200 to turn left. Additionally, or alternatively, control system 408 generates and transmits control signals to cause other devices (e.g., headlights, turn signal, door locks, windshield wipers, and / or the like) of vehicle 200 to change states.

[0068] In some embodiments, perception system 402, planning system 404, localization system 406, and / or control system 408 implement at least one machine learning model (e.g., at least one multilayer perceptron (MLP), at least one convolutional neural network (CNN), at least one recurrent neural network (RNN), at least one autoencoder, at least one transformer, and / or the like). In some examples, perception system 402, planning system 404, localization system 406, and / or control system 408 implement at least one machine learning model alone or in combination with one or more of the above-noted systems. In some examples, perception system 402, planning systemAttorney Docket No. 46154-0571WO1 / I2023191404, localization system 406, and / or control system 408 implement at least one machine learning model as part of a pipeline (e.g., a pipeline for identifying one or more objects located in an environment and / or the like). An example of an implementation of a machine learning model is included below with respect to FIGS. 4B-4D.

[0069] Database 410 stores data that is transmitted to, received from, and / or updated by perception system 402, planning system 404, localization system 406 and / or control system 408. In some examples, database 410 includes a storage component (e.g., a storage component that is the same as or similar to storage component 308 of FIG. 3) that stores data and / or software related to the operation and uses at least one system of autonomous vehicle compute 400. In some embodiments, database 410 stores data associated with 2D and / or 3D maps of at least one area. In some examples, database 410 stores data associated with 2D and / or 3D maps of a portion of a city, multiple portions of multiple cities, multiple cities, a county, a state, a State (e.g., a country), and / or the like). In such an example, a vehicle (e.g., a vehicle that is the same as or similar to vehicles 102 and / or vehicle 200) can drive along one or more drivable regions (e.g., single-lane roads, multi-lane roads, highways, back roads, off road trails, and / or the like) and cause at least one LIDAR sensor (e.g., a LIDAR sensor that is the same as or similar to LiDAR sensors 202b) to generate data associated with an image representing the objects included in a field of view of the at least one LiDAR sensor.

[0070] In some embodiments, database 410 can be implemented across a plurality of devices. In some examples, database 410 is included in a vehicle (e.g., a vehicle that is the same as or similar to vehicles 102 and / or vehicle 200), an autonomous vehicle system (e.g., an autonomous vehicle system that is the same as or similar to remote AV system 114, a fleet management system (e.g., a fleet management system that is the same as or similar to fleet management system 116 of FIG. 1, a V2I system (e.g., a V2I system that is the same as or similar to V2I system 118 of FIG. 1 ) and / or the like.

[0071] Referring now to FIG. 4B, illustrated is a diagram of an implementation of a machine learning model. More specifically, illustrated is a diagram of an implementation of a convolutional neural network (CNN) 420. For purposes of illustration, the following description of CNN 420 will be with respect to an implementation of CNN 420 by perception system 402. However, it will be understood that in some examples CNN 420Attorney Docket No. 46154-0571WO1 / I2023191(e.g., one or more components of CNN 420) is implemented by other systems different from, or in addition to, perception system 402 such as planning system 404, localization system 406, and / or control system 408. While CNN 420 includes certain features as described herein, these features are provided for the purpose of illustration and are not intended to limit the present disclosure.

[0072] CNN 420 includes a plurality of convolution layers including first convolution layer 422, second convolution layer 424, and convolution layer 426. In some embodiments, CNN 420 includes sub-sampling layer 428 (sometimes referred to as a pooling layer). In some embodiments, sub-sampling layer 428 and / or other subsampling layers have a dimension (i.e., an amount of nodes) that is less than a dimension of an upstream system. By virtue of sub-sampling layer 428 having a dimension that is less than a dimension of an upstream layer, CNN 420 consolidates the amount of data associated with the initial input and / or the output of an upstream layer to thereby decrease the amount of computations necessary for CNN 420 to perform downstream convolution operations. Additionally, or alternatively, by virtue of sub-sampling layer 428 being associated with (e.g., configured to perform) at least one subsampling function (as described below with respect to FIGS. 4C and 4D), CNN 420 consolidates the amount of data associated with the initial input.

[0073] Perception system 402 performs convolution operations based on perception system 402 providing respective inputs and / or outputs associated with each of first convolution layer 422, second convolution layer 424, and convolution layer 426 to generate respective outputs. In some examples, perception system 402 implements CNN 420 based on perception system 402 providing data as input to first convolution layer 422, second convolution layer 424, and convolution layer 426. In such an example, perception system 402 provides the data as input to first convolution layer 422, second convolution layer 424, and convolution layer 426 based on perception system 402 receiving data from one or more different systems (e.g., one or more systems of a vehicle that is the same as or similar to vehicle 102), a remote AV system that is the same as or similar to remote AV system 114, a fleet management system that is the same as or similar to fleet management system 116, a V2I system that is the same as or similar to V2I system 118,Attorney Docket No. 46154-0571WO1 / I2023191and / or the like). A detailed description of convolution operations is included below with respect to FIG. 4C.

[0074] In some embodiments, perception system 402 provides data associated with an input (referred to as an initial input) to first convolution layer 422 and perception system 402 generates data associated with an output using first convolution layer 422. In some embodiments, perception system 402 provides an output generated by a convolution layer as input to a different convolution layer. For example, perception system 402 provides the output of first convolution layer 422 as input to sub-sampling layer 428, second convolution layer 424, and / or convolution layer 426. In such an example, first convolution layer 422 is referred to as an upstream layer and sub-sampling layer 428, second convolution layer 424, and / or convolution layer 426 are referred to as downstream layers. Similarly, in some embodiments perception system 402 provides the output of sub-sampling layer 428 to second convolution layer 424 and / or convolution layer 426 and, in this example, sub-sampling layer 428 would be referred to as an upstream layer and second convolution layer 424 and / or convolution layer 426 would be referred to as downstream layers.

[0075] In some embodiments, perception system 402 processes the data associated with the input provided to CNN 420 before perception system 402 provides the input to CNN 420. For example, perception system 402 processes the data associated with the input provided to CNN 420 based on perception system 402 normalizing sensor data (e.g., image data, LiDAR data, radar data, and / or the like).

[0076] In some embodiments, CNN 420 generates an output based on perception system 402 performing convolution operations associated with each convolution layer. In some examples, CNN 420 generates an output based on perception system 402 performing convolution operations associated with each convolution layer and an initial input. In some embodiments, perception system 402 generates the output and provides the output as fully connected layer 430. In some examples, perception system 402 provides the output of convolution layer 426 as fully connected layer 430, where fully connected layer 430 includes data associated with a plurality of feature values referred to as F1, F2 . . . FN. In this example, the output of convolution layer 426 includes data associated with a plurality of output feature values that represent a prediction.Attorney Docket No. 46154-0571WO1 / I2023191

[0077] In some embodiments, perception system 402 identifies a prediction from among a plurality of predictions based on perception system 402 identifying a feature value that is associated with the highest likelihood of being the correct prediction from among the plurality of predictions. For example, where fully connected layer 430 includes feature values F1, F2,... FN, and F1 is the greatest feature value, perception system 402 identifies the prediction associated with F1 as being the correct prediction from among the plurality of predictions. In some embodiments, perception system 402 trains CNN 420 to generate the prediction. In some examples, perception system 402 trains CNN 420 to generate the prediction based on perception system 402 providing training data associated with the prediction to CNN 420.

[0078] Referring now to FIGS. 4C and 4D, illustrated is a diagram of example operation of CNN 440 by perception system 402. In some embodiments, CNN 440 (e.g., one or more components of CNN 440) is the same as, or similar to, CNN 420 (e.g., one or more components of CNN 420) (see FIG. 4B).

[0079] At step 450, perception system 402 provides data associated with an image as input to CNN 440 (step 450). For example, as illustrated, perception system 402 provides the data associated with the image to CNN 440, where the image is a greyscale image represented as values stored in a two-dimensional (2D) array. In some embodiments, the data associated with the image may include data associated with a color image, the color image represented as values stored in a three-dimensional (3D) array. Additionally, or alternatively, the data associated with the image may include data associated with an infrared image, a radar image, and / or the like.

[0080] At step 455, CNN 440 performs a first convolution function. For example, CNN 440 performs the first convolution function based on CNN 440 providing the values representing the image as input to one or more neurons (not explicitly illustrated) included in first convolution layer 442. In this example, the values representing the image can correspond to values representing a region of the image (sometimes referred to as a receptive field). In some embodiments, each neuron is associated with a filter (not explicitly illustrated). A filter (sometimes referred to as a kernel) is representable as an array of values that corresponds in size to the values provided as input to the neuron. In one example, a filter may be configured to identify edges (e.g., horizontal lines, verticalAttorney Docket No. 46154-0571WO1 / I2023191lines, straight lines, and / or the like). In successive convolution layers, the filters associated with neurons may be configured to identify successively more complex patterns (e.g., arcs, objects, and / or the like).

[0081] In some embodiments, CNN 440 performs the first convolution function based on CNN 440 multiplying the values provided as input to each of the one or more neurons included in first convolution layer 442 with the values of the filter that corresponds to each of the one or more neurons. For example, CNN 440 can multiply the values provided as input to each of the one or more neurons included in first convolution layer 442 with the values of the filter that corresponds to each of the one or more neurons to generate a single value or an array of values as an output. In some embodiments, the collective output of the neurons of first convolution layer 442 is referred to as a convolved output. In some embodiments, where each neuron has the same filter, the convolved output is referred to as a feature map.

[0082] In some embodiments, CNN 440 provides the outputs of each neuron of first convolutional layer 442 to neurons of a downstream layer. For purposes of clarity, an upstream layer can be a layer that transmits data to a different layer (referred to as a downstream layer). For example, CNN 440 can provide the outputs of each neuron of first convolutional layer 442 to corresponding neurons of a subsampling layer. In an example, CNN 440 provides the outputs of each neuron of first convolutional layer 442 to corresponding neurons of first subsampling layer 444. In some embodiments, CNN 440 adds a bias value to the aggregates of all the values provided to each neuron of the downstream layer. For example, CNN 440 adds a bias value to the aggregates of all the values provided to each neuron of first subsampling layer 444. In such an example, CNN 440 determines a final value to provide to each neuron of first subsampling layer 444 based on the aggregates of all the values provided to each neuron and an activation function associated with each neuron of first subsampling layer 444.

[0083] At step 460, CNN 440 performs a first subsampling function. For example, CNN 440 can perform a first subsampling function based on CNN 440 providing the values output by first convolution layer 442 to corresponding neurons of first subsampling layer 444. In some embodiments, CNN 440 performs the first subsampling function based on an aggregation function. In an example, CNN 440 performs the first subsamplingAttorney Docket No. 46154-0571WO1 / I2023191function based on CNN 440 determining the maximum input among the values provided to a given neuron (referred to as a max pooling function). In another example, CNN 440 performs the first subsampling function based on CNN 440 determining the average input among the values provided to a given neuron (referred to as an average pooling function). In some embodiments, CNN 440 generates an output based on CNN 440 providing the values to each neuron of first subsampling layer 444, the output sometimes referred to as a subsampled convolved output.

[0084] At step 465, CNN 440 performs a second convolution function. In some embodiments, CNN 440 performs the second convolution function in a manner similar to how CNN 440 performed the first convolution function, described above. In some embodiments, CNN 440 performs the second convolution function based on CNN 440 providing the values output by first subsampling layer 444 as input to one or more neurons (not explicitly illustrated) included in second convolution layer 446. In some embodiments, each neuron of second convolution layer 446 is associated with a filter, as described above. The filter(s) associated with second convolution layer 446 may be configured to identify more complex patterns than the filter associated with first convolution layer 442, as described above.

[0085] In some embodiments, CNN 440 performs the second convolution function based on CNN 440 multiplying the values provided as input to each of the one or more neurons included in second convolution layer 446 with the values of the filter that corresponds to each of the one or more neurons. For example, CNN 440 can multiply the values provided as input to each of the one or more neurons included in second convolution layer 446 with the values of the filter that corresponds to each of the one or more neurons to generate a single value or an array of values as an output.

[0086] In some embodiments, CNN 440 provides the outputs of each neuron of second convolutional layer 446 to neurons of a downstream layer. For example, CNN 440 can provide the outputs of each neuron of first convolutional layer 442 to corresponding neurons of a subsampling layer. In an example, CNN 440 provides the outputs of each neuron of first convolutional layer 442 to corresponding neurons of second subsampling layer 448. In some embodiments, CNN 440 adds a bias value to the aggregates of all the values provided to each neuron of the downstream layer. For example, CNN 440 adds aAttorney Docket No. 46154-0571WO1 / I2023191bias value to the aggregates of all the values provided to each neuron of second subsampling layer 448. In such an example, CNN 440 determines a final value to provide to each neuron of second subsampling layer 448 based on the aggregates of all the values provided to each neuron and an activation function associated with each neuron of second subsampling layer 448.

[0087] At step 470, CNN 440 performs a second subsampling function. For example, CNN 440 can perform a second subsampling function based on CNN 440 providing the values output by second convolution layer 446 to corresponding neurons of second subsampling layer 448. In some embodiments, CNN 440 performs the second subsampling function based on CNN 440 using an aggregation function. In an example, CNN 440 performs the first subsampling function based on CNN 440 determining the maximum input or an average input among the values provided to a given neuron, as described above. In some embodiments, CNN 440 generates an output based on CNN 440 providing the values to each neuron of second subsampling layer 448.

[0088] At step 475, CNN 440 provides the output of each neuron of second subsampling layer 448 to fully connected layers 449. For example, CNN 440 provides the output of each neuron of second subsampling layer 448 to fully connected layers 449 to cause fully connected layers 449 to generate an output. In some embodiments, fully connected layers 449 are configured to generate an output associated with a prediction (sometimes referred to as a classification). The prediction may include an indication that an object included in the image provided as input to CNN 440 includes an object, a set of objects, and / or the like. In some embodiments, perception system 402 performs one or more operations and / or provides the data associated with the prediction to a different system, described herein.

[0089] Referring now to FIG. 5, illustrated is a diagram of an implementation 500 of a process for construction zone polygon representation learning via pseudo-labels and symbiotic post-processing. In some embodiments, implementation 500 include a planning system 504a, a control system 504b, and a construction zone generator 504c. In some embodiments, planning system 504a is the same as or similar to planning system 404 of FIG. 4; control system 504b is the same as or similar to control system 408 of FIG. 4; andAttorney Docket No. 46154-0571WO1 / I2023191the construction zone generator is the same as or similar to a trained deep learning based model for construction zone segmentation as described with respect to FIGs. 7 and 8.

[0090] In examples, a planning system 504a generates (514) a route, and transmits (516) the route to the control system 504b. The control system 504b controls operation of the vehicle by generating and transmitting control signals to cause a powertrain, steering, or brake system to operate. In examples, a construction zone generator 504c identifies construction zones in the environment. The planning system 504a generates routes that avoid the construction zones as identified by the construction zone generator 504c.

[0091] Traditional approaches to identifying construction zones detect construction objects such as cones and barriers and apply rules to generate polygons. A rule based approach does not generalize well for complex construction zone scenarios. The construction zone generator 504c combines pseudo label generation and deep learning to directly predict construction zone polygons, addressing the limitations of traditional methods. The present systems and techniques employ a dual loss training framework that leverages both real and pseudo labels, enhanced by a symbiotic post-processing that ensures consistency between object detection and segmentation outputs. In examples, the overall loss is backpropagated along the construction zone segmentation pathway of the neural network. Through the use of pseudo labels, deep learning networks are trained to detect construction zones using available training data. Construction zones can occur infrequently in training data and thus data including construction zones can be limited.

[0092] FIG. 6 shows construction zone objects used to generate construction zone polygons or pseudo labels. In some embodiments, construction zone polygons or pseudo labels are used to predict construction zone segmentation masks. In examples, a construction zone segmentation mask is a pixel-level map that identifies which parts of an image belong to a construction zone. In FIG. 6, construction zone objects are shown by ovals. In examples, construction zone objects include temporary or semi-permanent objects that alter normal roadway conditions that are detected, classified, or otherwise accounted for by the vehicle. Construction zone objects include, for example, traffic control devices such as cones, delineator posts, channelizer drums, reflective bollards,Attorney Docket No. 46154-0571WO1 / I2023191temporary lane dividers, water-filled barriers, concrete barriers, and portable rumble strips. Additional construction zone objects include construction related signage such as portable warning signs, detour indicators, lane-closure markers, speed-limit adjustments, arrow boards, and variable message displays. The construction zone objects also include construction personnel, including workers wearing high-visibility apparel and flaggers operating stop / slow paddles. Construction zone objects further include construction machinery and vehicles such as excavators, backhoes, bulldozers, dump trucks, rollers, graders, pavers, skid steers, and utility trucks. Equipment and materials designated as construction zone objects can also include generators, compressors, lighting towers, tool carts, jackhammers, air hoses, cables, gravel piles, dirt mounds, asphalt materials, steel plates, rebar bundles, lumber stacks, and sandbags. Moreover, temporary roadway elements classified as construction zone objects may include tape-based or painted lane markings, raised pavement markers, fresh asphalt patches, milled pavement, trench plates, and temporary curbs. The construction zone objects may also include fencing or barriers such as safety mesh, temporary chain-link fencing, and plastic barricade systems, as well as miscellaneous items including warning flags, toolboxes, portable toilets, and worksite trailers. In examples, construction zone objects represent the dynamic and heterogeneous conditions characteristic of construction zones and inform perception, planning, and control systems of a vehicle. For ease of explanation, construction zone objects are described as traffic cones and barriers. However, any construction zone objects can be used in accordance with the present systems and techniques.

[0093] In the example of FIG. 6, a distance-based rule-based approach is used to create construction zone polygons 624, 626, and 628 with pseudo labels. A vehicle 612 is shown on a map 610 traversing a roadway with a construction zone. In examples, the map 610 is an instance of a localization layer of a high-definition (HD) map. The corresponding polygons are shown on a map 620, which in examples is an instance of a road geometry layer of an HD map. The creation of pseudo labels is fast, with a runtime of about 0.01 seconds using optimized Python code. In examples, construction zone objects are identified in the environment. Distances between the construction zone objects are calculated, and if the distance between objects exceeds a threshold, they areAttorney Docket No. 46154-0571WO1 / I2023191not included in the same construction zone polygon. Based on the construction zone objects and their respective distances, construction zone polygons are created. These polygons are used to identify construction zones in the environment and are annotated using pseudo labels. The construction zone polygons are generated using rules and heuristics.

[0094] In examples, the data used to generate pseudo labels of construction zones is based on sensor data, such as camera, LiDAR, radar, or any combinations thereof. The coordinates of the construction zone and the size of the construction zone are generated from available sensor data. A list of locations of the objects is created by annotating the data.

[0095] In examples, the generated construction zone polygons are used to mine for complex construction zones. Pseudo labels can include information identifying the complexity of construction zones. Complex construction zones are mined and then annotated manually. Rather than using manual annotation resources to review a large number of different logs, the present systems and techniques identify those particular complex scenes for manual annotation, reducing the workload on manual annotators. In examples, the complexity of a construction zone is related to the number of construction objects on the road and the size of the construction zone. In examples, the number of edges of the construction zone polygon can indicate the complexity of the construction zone. For example, as the number of edges of a construction zone polygon increases, the complexity of the construction zone increases. Further, the shape of the construction zone polygon indicates the complexity of the construction zone. For example, while a trapezoid and a rectangle have the same number of sides, a trapezoid shaped construction zone may be considered a more complex construction zone. The four angles of a rectangle are right angles (90°), and opposite sides are equal and parallel. By contrast, one pair of opposite sides is parallel for a trapezoid, and the other sides can be of different lengths. The angles of a trapezoid are not necessarily 90° and can vary, making the shape less uniform and more complex than a rectangle.

[0096] FIG. 7 shows training a deep learning based model for construction zone segmentation. Sensor data 702 from the environment is obtained. In examples, the sensor data includes lidar data, camera data, and radar data. In examples, the sensorAttorney Docket No. 46154-0571WO1 / I2023191data is used to generate pseudo-labels. The output of pseudo label generation is a first set of construction zone polygons labeled with pseudo labels, and a second set of construction zone polygons labeled by human-expert annotators.

[0097] The sensor data 702 is input to a neural network 704. The neural network 704 is communicatively coupled with a detection head 706 and a construction zone segmentation head 708. The detection head 706 outputs detected construction zone objects 710. The construction zone segmentation head 708 outputs a construction zone segmentation 712. In some embodiments, sensor data input to the neural network is manually annotated to identify construction zone objects for training.

[0098] The detection head 706 converts high-level spatial features into discrete object detections. In examples, the detection head 706 operates on BEV feature maps produced by the neural network 704, and includes a set of lightweight convolutional branches that independently regress object geometry and generate object classifications. In some embodiments, the detection head 706 outputs dense predictions over a BEV grid, which are subsequently filtered using confidence thresholds and non-maximum suppression to generate a final set of object detections (e.g., detected construction zone objects 710). In examples, particular classifications are associated with construction zones.

[0099] The construction zone segmentation head 708 identifies the pixels in a bird's eye view (BEV) that are associated with a construction zone. In examples, a BEV is a top-down perspective of the environment. Additionally, in examples, a pixel in BEV corresponds to a 0.15 meter by 0.15 meter grid-cell in BEV. For example, a pixel in BEV corresponds to a grid-cell that is part of a discretized coordinate system representing a ground plane. In some embodiments, a grid-cell represents a semantic class of the corresponding ground region in the environment. In examples, the semantic class segments the environment into locations that are a construction zone or not a construction zone.

[0100] In some embodiments, BEV semantic segmentation is implemented using knowledge distillation by training a lightweight student network to mimic the output structure and spatial reasoning of a larger, high-capacity teacher model. During training, both models process the same BEV inputs, and the student is supervised using twoAttorney Docket No. 46154-0571WO1 / I2023191complementary signals including a segmentation loss computed against ground-truth BEV labels, and a distillation loss that biases the student to match the teacher’s soft per-pixel class distributions. In examples, the teacher’s logits are softened using a temperature parameter to obtain inter-class relationships (e.g., construction zone or not a construction zone). For example, output probabilities of a model are made less sharp and more informative by dividing its logits by a temperature before applying softmax. In the knowledge distillation framework, a hard segmentation loss is combined with a KL-divergence term between teacher and student logits, optionally augmented with feature-level distillation if intermediate BEV feature maps are available. In examples, intermediate BEV feature maps are the hidden representations produced within a model and are internal feature tensors the model builds while reasoning over inputs. Exposing intermediate BEV feature maps enables knowledge distillation at the feature level. Through knowledge distillation, the student inherits the teacher’s spatial priors and class-boundary structure, improving segmentation quality while remaining efficient enough for real-time BEV perception.

[0101] In examples, the detected construction zone objects 710 and the construction zone segmentation 712 are used to compute one or more losses at block 716. In examples, a first loss is computed based on real annotations, if available. A second loss is computed based on pseudo labels from ground truth object construction zone objects. In some embodiments, the losses are implemented in a knowledge distillation framework via symbiotic post-processing. The knowledge distillation framework transfers knowledge from a large, complex model, sometimes referred to the teacher, to a smaller, more efficient model sometimes referred to as the student. This process enables the student model to achieve similar performance to the teacher model while being less computationally intensive. Symbiotic post-processing is described with respect to FIGS. 9-10.

[0102] FIG. 8 shows the architecture of the deep learning based model for construction zone segmentation. As shown in the example of FIG. 8, the construction zone segmentation head is configured with multiple layers (deep) but with each layer having a relatively small number of neurons (thin). In examples, fewer neurons per layer can yield models that are easier to train and less prone to overfitting. Moreover, theAttorney Docket No. 46154-0571WO1 / I2023191construction zone segmentation head is computationally efficient, requiring less memory and processing power. The construction zone segmentation head includes a 128 channel input feature size. Seven layers are used to convolve the input features into two channels that identify if a respective pixel corresponds to a construction zone or does not correspond to a construction zone.

[0103] As shown in the example of FIG. 8, the construction zone segmentation head 708 starts with an input feature map of size 128 channels. The feature map is output from a backbone neural network 704. The neural network 704 can include a feature extractor in a convolutional neural network (CNN). The construction zone segmentation head 708 converts the 128-channel input features into a 2-channel output, where each channel represents a class for pixel-wise classification. In the example of FIG. 8, the construction zone segmentation head 708 consists of seven convolutional layers. Each layer applies a convolution operation to transform the input features. The construction zone segmentation head 708 gradually reduces the number of channels while preserving spatial information. After passing through the seven convolutional layers, the final output is a 2-channel feature map. Each pixel in this map is classified into one of two classes based on the values in these two channels as being a construction zone or not a construction zone. The construction zone segmentation 712 is a per pixel identification of a construction zone. In some embodiments, training includes generating a rule based construction zone at block 804 using construction zone objects (e.g., objects associated with construction zones). At block 806, symbiotic post-processing is used to improve the generation of the rule based construction zones based on detected objects and the construction zone segmentation head. For example, traffic cones (e.g., detected construction zone objects) that are more than a predetermined distance (e.g., five meters) from the corresponding segmented construction zone (e.g., construction zone segmentation) are assigned reduced confidence scores. Second, construction zone segments that are beyond the threshold distance from construction zone objects are discarded. For example, if a corresponding segmented construction zone is beyond a threshold distance (e.g., five meters radius) from detected construction zone objects, the segment is discarded. In some embodiments, the symbiotic post-processing is performed iteratively, such that frames of sensor data are iteratively evaluated to determineAttorney Docket No. 46154-0571WO1 / I2023191confidence scores and exclude construction zone segments that the beyond a predetermined radius of a construction zone object.

[0104] Reference number 802 shows a particular structure of a lightweight convolutional segmentation head that refines high-level BEV features while preserving spatial resolution to generate per-pixel semantic predictions (e.g., a construction zone or does not correspond to a construction zone) in BEV space. The segmentation head is configured to obtain a 128-channel BEV feature map and processes it through a sequence of seven uniform refinement blocks. Each block consists of batch normalization, a ReLU activation, and a 3×3 convolution. The first block performs channel reduction, projecting the 128-channel input into a 16-channel representation. Blocks 2 through 6 maintain this 16-channel width, applying repeated 3×3 convolutions to progressively refine spatial structure and expand the effective receptive field without altering the BEV grid resolution. The final block maps the refined 16-channel representation to a 2-channel output, producing dense per-cell logits corresponding to the target semantic classes. The segmentation head is efficient with low-latency that enables real-time BEV perception.

[0105] FIG. 9 shows the architecture of the deep learning based model for construction zone segmentation with iterative symbiotic post-processing 902. The symbiotic post-processing 902 uses the rule based construction zone at block 804 and the construction zone segmentation 712 to improve each of the identified construction zones. In the example of FIG. 9, at block 904 the rule based construction zone at block 804 is used to inform a confidence score of construction zone objects that are beyond a threshold distance from a construction zone identified in the construction zone segmentation 712. In examples, a confidence score is first computed as a direct output of the neural network. The symbiotic post-processing 902 reduces a confidence score of traffic cones that are far from segmented construction zones. At block 906, if a change in segmentation satisfies a predetermined threshold, symbiotic post-processing returns to block 804. Accordingly, a construction zone determined based on one or more rules is used to iteratively inform a confidence score associated with construction zone objects that do not satisfy a predetermined distance with respect to a construction zone identified via construction zone segmentation output by a trained machine learning model.Attorney Docket No. 46154-0571WO1 / I2023191

[0106] In examples, a change in segmentation causes the predicted class for some pixels to change based on the change in confidence scores at block 904. When the change in segmentation satisfies a predetermined threshold, process flow returns to block 804 where a construction zone is generated based on a set of one or more rules using construction zone objects. The rule based construction zone at block 804 and confidence scores at block 904 are iteratively determined based on the resulting changes in segmentation. Consider an example of a flickering traffic cone. In this example, the traffic cone is not actually present, however a trained object detection head intermittently observes the traffic cone, which may actually be a small pillar or other object in the environment. In this example, an output of the segmentation head (e.g., construction zone segmentation) predicts that the construction polygon ends prior to the intermittently observed the traffic cone (e.g., detected construction zone object). In this example, the misidentified traffic cone is discarded based on a reduction in confidence score and an evaluation of a change in segmentation resulting from the flickering traffic cone.

[0107] FIG. 10 shows the calculation of losses with and without knowledge distillation. In examples, the losses represent a realignment of a detection head (e.g., detection head 706 of FIG. 7) with a construction zone segmentation head (e.g., construction zone segmentation head 708 of FIG. 7).

[0108] A flowchart at reference number 1005 shows the calculation of loss without knowledge distillation. At block 1010, it is determined if a real annotation is available. In examples, the real annotation is a manual annotation generated by a human. In some embodiments, the real annotation is a per pixel annotation. If a real annotation is available, the loss is computed based on the manual annotation of the construction zone at block 1012. If a real annotation is not available, the loss is computed based on the pseudo-label construction zone that is generated based on ground-truth construction zone objects, such as traffic cones, barriers, and the like at block 1014.

[0109] A flowchart at reference number 1015 shows the calculation of loss with knowledge distillation. At block 1020, it is determined if a real annotation is available. If a real annotation is available, the loss is computed based on the manual annotation of the construction zone at block 1022. A knowledge distillation framework is applied, where the total training objective is expressed as Total loss = Loss1+ w·Loss2, which reflects theAttorney Docket No. 46154-0571WO1 / I2023191two complementary sources of supervision used to train the student model. Loss1represents a task loss computed against the ground-truth labels. For example, the cross-entropy loss is used in segmentation to ensure the student correctly predicts the class of each pixel. Loss2captures the distillation component. For example, Loss2measures how closely the student matches the teacher model’s behavior. This measurement may be determined through a KL-divergence term comparing the student's softened logits to those of the teacher. The scalar w controls the relative importance of teacher guided learning where the student model learns from the ground-truth labels and / or directly from the teacher model’s behavior. Accordingly, the training process balances direct supervision from the labeled data with the richer, more nuanced information encoded in the teacher’s output distribution. Together, the losses enable the student to learn both the correct labels and the teachers internal understanding of class relationships, leading to a more capable and efficient model.

[0110] Referring again to reference number 1015, if a real annotation is not available, the loss is computed based on the pseudo-label construction zone that is generated based on ground-truth construction zone objects, such as traffic cones, barriers, and the like at block 1024. When used with symbiotic post-processing, the first loss and second loss are summed to determine a total loss used in the knowledge distillation framework.

[0111] The symbiotic post-processing improves the detection of the construction zone objects. The present systems and techniques improve data efficiency, as a large dataset with construction zones may be unavailable. Automated pseudo label generation reduces annotation cost. The present systems and techniques can be used to intelligently mine complex cases and target those cases for human annotation. Accordingly, the present systems and techniques enable continuous system improvement through edge case discovery. This provides a more scalable method for continuous improvement enabling coverage of unique or outlier cases, as the rules are not updated or changed to such cases.

[0112] Referring now to FIG. 11, illustrated is a flowchart of a process 1100 for construction zone polygon representation learning via pseudo-labels and symbiotic post¬ processing. In some embodiments, one or more of the steps described with respect toAttorney Docket No. 46154-0571WO1 / I2023191process 1100 are performed (e.g., completely, partially, and / or the like) by autonomous system 202. Additionally, or alternatively, in some embodiments one or more steps described with respect to process 1100 are performed (e.g., completely, partially, and / or the like) by another device or group of devices separate from or including autonomous system 202 such as device 300 of FIG. 3.

[0113] At block 1102, a machine learning model is trained to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iteratively updating multiple loss functions until a consensus is reached between the multiple loss functions. In examples, an overall loss is backpropagated along the construction zone segmentation pathway of the neural network. A first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data. In some embodiments, the pseudo-labeled data identifies a complexity of the construction zone and size of a construction zone in the samples of driving data.

[0114] In examples, at least one loss function is applied to the samples of driving data annotated by humans (e.g., expert annotated data). Human annotated training data is obtained by selecting samples of driving data for human annotation based on a complexity or size of a construction zone identified in the pseudo-labeled data. Accordingly, edge cases can be selected based on complexity or size as identified in the pseudo-labeled data.

[0115] At block 1104, the construction zone object annotations are aligned with the predicted construction zone segmentation masks. In some embodiments, predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded. In examples, predicted construction zone segmentation masks remaining after discarding are used to finetune the trained machine learning model.

[0116] Embodiments

[0117] According to some non-limiting embodiments or examples, provided is a method. The method includes training a machine learning model to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iterativelyAttorney Docket No. 46154-0571WO1 / I2023191updating multiple loss functions until a consensus is reached between the multiple loss functions, wherein a first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data. The method also includes aligning the construction zone object annotations with the predicted construction zone segmentation masks, wherein a portion of predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded.

[0118] According to some non-limiting embodiments or examples, provided is a system. The system includes at least one processor. The system also includes at least one non-transitory storage media storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations. The operations include training a machine learning model to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iteratively updating multiple loss functions until a consensus is reached between the multiple loss functions, wherein a first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data. The operations also include aligning the construction zone object annotations with the predicted construction zone segmentation masks, wherein a portion of predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded.

[0119] According to some non-limiting embodiments or examples, provided is at least one non-transitory storage media storing instructions that, when executed by at least one processor, cause the at least one processor to perform operations. The operations include training a machine learning model to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iteratively updating multiple loss functions until a consensus is reached between the multiple loss functions, wherein a first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data. The operations alsoAttorney Docket No. 46154-0571WO1 / I2023191include aligning the construction zone object annotations with the predicted construction zone segmentation masks, wherein a portion of predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded.

[0120] Further non-limiting aspects or embodiments are set forth in the following numbered embodiments:

[0121] Embodiment 1: A method, comprising: training a machine learning model to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iteratively updating multiple loss functions until a consensus is reached between the multiple loss functions, wherein a first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data; and aligning the construction zone object annotations with the predicted construction zone segmentation masks, wherein a portion of predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded.

[0122] Embodiment 2: The method of any preceding embodiments, wherein the pseudo-labeled data identifies a complexity and size of a construction zone in samples of driving data.

[0123] Embodiment 3: The method of any preceding embodiments, wherein at least one loss function is applied to samples of driving data annotated by humans.

[0124] Embodiment 4: The method of any preceding embodiments, wherein hu¬ man annotated training data is obtained by selecting samples of driving data for human annotation based on a complexity or size of a construction zone identified in the pseudo¬ labeled data.

[0125] Embodiment 5: The method of any preceding embodiments, wherein at least one loss function is applied to the pseudo-labeled data.

[0126] Embodiment 6: The method of any preceding embodiments, wherein predicted construction zone segmentation masks remaining after discarding the portion are used to finetune the trained machine learning model.Attorney Docket No. 46154-0571WO1 / I2023191

[0127] Embodiment 7: A system comprising at least one processor; and at least one non-transitory storage media storing instructions that, when executed by the at least one processor, cause the at least one processor perform the method of any of embodi¬ ments 1 to 6.

[0128] Embodiment 8: At least one non-transitory storage media storing instructions that, when executed by at least one processor, cause the at least one processor perform the method of any of embodiments 1 to 6.

[0129] In the foregoing description, aspects and embodiments of the present disclosure have been described with reference to numerous specific details that can vary from implementation to implementation. Accordingly, the description and drawings are to be regarded in an illustrative rather than a restrictive sense. The sole and exclusive indicator of the scope of the invention, and what is intended by the applicants to be the scope of the invention, is the literal and equivalent scope of the set of claims that issue from this application, in the specific form in which such claims issue, including any subsequent correction. Any definitions expressly set forth herein for terms contained in such claims shall govern the meaning of such terms as used in the claims. In addition, when we use the term “further comprising,” in the foregoing description or following claims, what follows this phrase can be an additional step or entity, or a sub-step / sub-entity of a previously-recited step or entity.

Claims

1. Attorney Docket No. 46154-0571WO1 / I2023191WHAT IS CLAIMED IS:

1. A method, comprising:training a machine learning model to predict construction zone segmentation masks in an environment based on pseudo-labeled data by iteratively updating multiple loss functions until a consensus is reached between the multiple loss functions, wherein a first loss function updates parameters of the machine learning model to label pixels of driving data as a construction zone or not a construction zone and a second loss function is based on annotated construction zone objects in the driving data; andaligning construction zone object annotations with the predicted construction zone segmentation masks, wherein a portion of predicted construction zone segmentation masks greater than a threshold distance from a construction zone object annotation are discarded.

2. The method of claim 1, wherein the pseudo-labeled data identifies a complexity and size of a construction zone in samples of driving data.

3. The method of claim 1, wherein at least one loss function is applied to samples of driving data annotated by humans.

4. The method of claim 1, wherein human annotated training data is obtained by selecting samples of driving data for human annotation based on a complexity or size of a construction zone identified in the pseudo-labeled data.

5. The method of claim 1, wherein at least one loss function is applied to the pseudo-labeled data.

6. The method of claim 1, wherein predicted construction zone segmentation masks remaining after discarding the portion are used to finetune the trained machine learning model.Attorney Docket No. 46154-0571WO1 / I20231917. A system comprising:at least one processor; andat least one non-transitory storage media storing instructions that, when executed by the at least one processor, cause the at least one processor perform the method of any of claims 1 to 6.

8. At least one non-transitory storage media storing instructions that, when executed by at least one processor, cause the at least one processor perform the method of any of claims 1 to 6.

Images