Error mining system

The error mining system addresses inefficiencies in autonomous vehicle error detection by automatically generating datasets for model training, enhancing annotation efforts and improving navigation accuracy.

WO2026236062A1PCT designated stage Publication Date: 2026-11-12MOTIONAL AD LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/US2026/026321
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-06
Filing Date
2026-05-01
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Autonomous vehicles face challenges in efficiently identifying and correcting errors in sensor data, leading to inaccurate decision-making and navigation, which current annotation methods are time-consuming and expensive.

Method used

An error mining system that automatically mines data for errors in unlabeled data, generating datasets relevant to the subject modality, improving autonomous system functionality by focusing annotation efforts on identified errors and enhancing model training.

Benefits of technology

Enhances the efficiency and effectiveness of data annotation by proactively identifying errors, increasing recall, and addressing long-tail problems in autonomous driving, thereby improving the accuracy and reliability of autonomous vehicle systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2026026321_12112026_PF_FP_ABST
    Figure US2026026321_12112026_PF_FP_ABST
Patent Text Reader

Abstract

Provided are methods for an error mining system, which can include generating attributes based on a subject modality and modeling framework; extracting records from driving data according to the generated attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; and evaluating the subject modality within the modeling framework. Systems and computer program products are also provided.
Need to check novelty before this filing date? Find Prior Art

Description

Attorney Docket No. 46154-0577WO1 / I2023201Error Mining SystemCROSS REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to Greek Patent Application No.20250100331 filed on May 6, 2025, the entire contents of which are hereby incorporated by reference.BACKGROUND

[0002] Autonomous vehicles sense the surrounding environment and navigate independently. Autonomous vehicles rely on multiple sensors to perceive the environment. The sensors capture data representative of the surrounding environment, and the autonomous vehicle processes the data to make decisions. These decisions enable the autonomous vehicle to choose a path that avoids obstacles and reacts to a variety of different driving scenarios.BRIEF DESCRIPTION OF THE FIGURES

[0003] FIG. 1 is an example environment in which a vehicle including one or more components of an autonomous system can be implemented;

[0004] FIG. 2 is a diagram of one or more systems of a vehicle including an autonomous system;

[0005] FIG. 3 is a diagram of components of one or more devices and / or one or more systems of FIGS. 1 and 2;

[0006] FIG. 4A is a diagram of certain components of an autonomous system;

[0007] FIG. 4B is a diagram of an implementation of a neural network;

[0008] FIG. 4C and 4D are a diagram illustrating example operation of a CNN;

[0009] FIG. 5 is a diagram of an implementation of an error mining system.

[0010] FIG. 6 is workflow implemented at an error mining system.

[0011] FIG. 7 shows a false positive in LiDAR based object detection and segmentation.

[0012] FIG. 8 shows a false negative camera based object detection.Attorney Docket No. 46154-0577WO1 / I2023201

[0013] FIG. 9 shows a false positive of camera based object detection.

[0014] FIG. 10 shows a false negative in LiDAR based object detection.

[0015] FIG. 11 shows a false positive vehicle detection by a LiDAR-based object detection model.

[0016] FIGs. 12A and 12B show flickering predictions from a semantic model.

[0017] FIG. 13 shows inaccurate predictions from a prediction model.

[0018] FIG. 14 shows inaccurate heading / flickering by a LiDAR-based object detection model.

[0019] FIG. 15 is a process that enables an error mining system.DETAILED DESCRIPTION

[0020] 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.

[0021] 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.

[0022] 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.Attorney Docket No. 46154-0577WO1 / I2023201In 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 can represent one or multiple signal paths (e.g., a bus), as may be needed, to affect the communication.

[0023] 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.

[0024] 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.

[0025] 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,Attorney Docket No. 46154-0577WO1 / I2023201instructions, 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 the information 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.

[0026] 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.

[0027] 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, andAttorney Docket No. 46154-0577WO1 / I2023201networks have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0028] General Overview

[0029] In some aspects and / or embodiments, systems, methods, and computer program products described herein include and / or implement an error mining system. A vehicle (such as an autonomous vehicle) can include computing systems that execute machine learning models to enable various vehicle functions. The machine learning models are trained using datasets that include driving data. Attributes are generated based on a specified configuration of a vehicle. Records are extracted from the driving data based on the attributes. A dataset is generated by aggregating the records into a predetermined format. The subject modality is evaluated using the generated dataset.

[0030] By virtue of the implementation of systems, methods, and computer program products described herein, techniques for an error mining system enable fast and efficient mining for data used to train machine learning models. Some of the advantages of these techniques include a single framework that automatically mines data for errors in unlabeled data and generates a dataset relevant to the subject modality. Autonomous system functionality is improved according to the features of the single framework that automatically mines data for errors. The present techniques enable the generation of datasets corresponding to distinct and independent autonomy functions, where the datasets are used to train, validate or finetune respective models. By performing error mining prior to annotation, data annotation is focused on those identified errors, maximizing the value of annotation efforts on improving the model. Furthermore, this scalable method is adaptable to raw datasets, increases recall and proactively finds potential errors, enabling more effective solutions to long-tail problems like autonomous driving. The framework proactively extracts records associated with errors, prior to the errors being discovered through time consuming and expensive labeling.

[0031] 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 systemAttorney Docket No. 46154-0577WO1 / I2023201118. 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 12, 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.

[0032] 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).

[0033] 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.

[0034] 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 secondAttorney Docket No. 46154-0577WO1 / I2023201spatiotemporal 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 106 include 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.

[0035] 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.

[0036] Vehicle-to-lnfrastructure (V2I) device 110 (sometimes referred to as a Vehicle-to-lnfrastructure or Vehicle-to-Everything (V2X) device) includes at least oneAttorney Docket No. 46154-0577WO1 / I2023201device 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 to communicate 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.

[0037] 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.

[0038] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0039] 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).

[0040] In some embodiments, V2I system 118 includes at least one device configured to be in communication with vehicles 102, V2I device 110, remote AV system 114, 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).

[0041] 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.

[0042] 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, fullyAttorney Docket No. 46154-0577WO1 / I2023201autonomous 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 support features. 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.

[0043] 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.

[0044] Cameras 202a include at least one device configured to be in communication with communication device 202e, autonomous vehicle compute 202f,Attorney Docket No. 46154-0577WO1 / I2023201and / 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 such an 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.

[0045] 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 orAttorney Docket No. 46154-0577WO1 / I2023201more 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.

[0046] 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. In 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.

[0047] 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 byAttorney Docket No. 46154-0577WO1 / I2023201some 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.

[0048] Microphones 202d includes 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). 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.

[0049] 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).

[0050] 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,Attorney Docket No. 46154-0577WO1 / I2023201graphical 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 110 of 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).

[0051] 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.

[0052] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0053] 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 / or the 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.

[0054] 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.

[0055] 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.

[0056] 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 beAttorney Docket No. 46154-0577WO1 / I2023201located in the near side of vehicle 200 in FIG. 2, brake system 208 may be located anywhere in vehicle 200.

[0057] 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) and / 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.

[0058] 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.

[0059] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0060] 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).

[0061] In some embodiments, communication interface 314 includes a transceiver¬ like component (e.g., a transceiver, a separate receiver and transmitter, and / or the like) that 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.

[0062] 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.

[0063] 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 processesAttorney Docket No. 46154-0577WO1 / I2023201described herein. Thus, embodiments described herein are not limited to any specific combination of hardware circuitry and software unless explicitly stated otherwise.

[0064] 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.

[0065] 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, the term “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.

[0066] 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.

[0067] 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,Attorney Docket No. 46154-0577WO1 / I2023201control 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), computer hardware (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).

[0068] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0069] 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 404 receives 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.

[0070] 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 physicalAttorney Docket No. 46154-0577WO1 / I2023201properties (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.

[0071] 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 associated with 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.

[0072] 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 angleAttorney Docket No. 46154-0577WO1 / I2023201of 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.

[0073] 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 system 404, 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.

[0074] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0075] 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.

[0076] 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 420 (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.

[0077] 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.Attorney Docket No. 46154-0577WO1 / I2023201

[0078] 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, and / or the like). A detailed description of convolution operations is included below with respect to FIG. 4C.

[0079] 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.

[0080] 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 associatedAttorney Docket No. 46154-0577WO1 / I2023201with 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).

[0081] 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.

[0082] 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.

[0083] 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).

[0084] 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 aAttorney Docket No. 46154-0577WO1 / I2023201color 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.

[0085] 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, vertical lines, 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).

[0086] 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.

[0087] 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 toAttorney Docket No. 46154-0577WO1 / I2023201corresponding 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.

[0088] 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 subsampling function 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.

[0089] 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.

[0090] 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 moreAttorney Docket No. 46154-0577WO1 / I2023201neurons 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.

[0091] 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 a bias 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.

[0092] 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.

[0093] 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 predictionAttorney Docket No. 46154-0577WO1 / I2023201(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.

[0094] In certain embodiments, computing systems and machine learning models undergo error mining to identify issues within the systems and models. This process involves collecting various types of data, including logs, sensor data, and environmental data. Errors are detected and categorized based on their characteristics in relation to the collected data. The root causes of the detected errors are then analyzed using statistical techniques, machine learning, manual review, or any combinations thereof. Finally, solutions are implemented to address the identified errors.

[0095] Referring now to FIG. 5, illustrated is a diagram of an error mining system 500. In examples, the error mining system 500 generates data used to train a machine learning model, such as the machine learning model described with respect to FIGs. 4A-4D. The error mining system includes an attribute generator 502 and a dataset generator 510. A column oriented database 504 stores driving data, such as driving data logged during the operation of a vehicle. Driving data can include, for example, AV driving data, human driving data, or any combinations thereof. In examples, the driving data is stored in the column oriented database 504 in a columnar storage file format optimized for large- scale data processing. For example, the driving data is stored in Parquet format. The column oriented database enables compression and efficient storage of driving data, which typically results in large datasets comprising sensor data, images, video and the like. In some embodiments, the column oriented database 504 enables extraction of specific datapoints without scanning the entire database. Further, the column oriented database 504 stores per-agent records in a hierarchical format, including nested sensor readings and metadata.

[0096] An attribute generator 502 extracts records from the column oriented database 504 based on attributes generated according to mining specifications 506. In some embodiments, attributes are properties or characteristics that are used to describe and analyze the behavior of a system or function. The properties or characteristics areAttorney Docket No. 46154-0577WO1 / I2023201determined from data corresponding to the system or function. In examples, custom attributes are tailored to specific needs based on mining specifications, which enables more efficient identification of errors in the system or function.

[0097] In examples, the mining specifications include a configuration used to evaluate a system. The configuration can include rules that describe correct behavior and error conditions and mining parameters that guide the mining process. For example, the configuration may be in the form of a configuration file specified by a user that identifies conditions associated with errors within a system. The configuration file can include, for example, parameters or settings associated with training, validating, or finetuning of a machine learning model. In some embodiments, the attribute generator selects records or agents-of-interest from the column oriented database 504 based on pre-filtering and / or cross-modality evaluation results, and computes custom attributes for those agents. In examples, pre-filtering is based on parameters in the configuration file, such as parameters associated with a time period, location, or specific events. The attribute generator 502 analyzes data in the column oriented database across different modalities (e.g., radar, LiDAR, camera) to identify agents of interest. Custom attributes are generated based on the mining specifications 506.

[0098] The attribute generator outputs records extracted from the column oriented database 504 based on attributes generated according to mining specifications 506. At block 508, a refinement may be applied to the records. If the records output by the attribute generator are to be refined, mining queries are obtained from the mining specifications 506. If the refinement is unnecessary, the records output by the attribute generator are input to a dataset generator 510. In some embodiments, the records are refined based on patterns observed in the extracted records.

[0099] The dataset generator 510 obtains records and generates a custom format dataset 512. For example, the dataset generator generates a scene-level metadata database, a custom-made Parquet database, or other format database. The scene-level metadata database includes information about the environment, such as weather conditions, time of day, and location. In examples, the dataset generator 510 ingests output of the attribute generator 502. For example, based on a user specified filter query, the dataset generator 510 completes a sequence of tasks such as running the query toAttorney Docket No. 46154-0577WO1 / I2023201obtain a subset of agent records. The dataset generator 510 aggregates the discrete, timestamp-based records into intervals. The generated intervals of data trigger cutlog generation, auto-labeling, or labeling by a human annotator. The generated data is transformed into a custom format such as a scene multimedia database (MMDB), custom-made parquet databases, and the like.

[0100] Using the error mining system, errors are mined in unlabeled datasets, extending error mining capabilities beyond the traditional comparison of model output to ground truth patterns. Traditional techniques mine errors based on annotated data. The annotated data identifies a correct / right answer or the ground truth. Traditionally, the ground truth is identified using human or expert manual labeling, and the output of a model is compared with the ground truth. Errors or other mistakes by the model are identified based on the ground truth. These traditional methods consume large amounts of annotation efforts prior to mining. These traditional methods are not scalable, not efficient, and expensive. Additionally, error discovery using annotated datasets is reactive. By mining data post incident, when the data has been annotated, errors are discovered only after they have been identified.

[0101] The mining specifications described herein introduce multiple ground truth-free approaches such as cross-modality, temporal-based, and prior / context-based mining. Moreover, the error mining system is highly scalable and adaptable to raw, incoming datasets, significantly boosting recall. The error mining system performs error discovery first, prior to annotation. The records identified by the attribute generator are annotated by the dataset generator. The error mining system proactively discovers potential errors, enabling resolution of system issues where a small number of events are common while a large number of events are rare.

[0102] The present techniques include a multipurpose and self-serviceable error mining system. The error mining system includes components that facilitate various modeling frameworks, including both traditional output (e.g., a comparison against the ground truth) and ground truth free approaches such as cross model training, temporal training, and prior training. The error mining system can mine multiple different error cases for multiple distinct, independent models without code changes or new workflowAttorney Docket No. 46154-0577WO1 / I2023201implementations. The error mining system adapts to new capabilities on top of existing ones.

[0103] Referring now to FIG. 6, a workflow 600 implemented at an error mining system is shown. In examples, the error mining system is the same as or similar to the error mining system 500 of FIG. 5. In examples, the attribute generator 606 and dataset generator 610 are the same as or similar to the attribute generator 502 and dataset generator 510 of FIG. 5. Mining specifications 602 include a specified configuration 604 and mining queries 614. In some embodiments, the specified configurations 604 are used to develop attributes that correspond to a modeling framework used to analyze the subject modality based on the generated datasets.

[0104] In examples, the configuration 604 specifies particular modules, systems, and / or functions to be evaluated using the generated datasets. For example, the subject module, system, or function includes components of an autonomous system, such as the autonomous system 202 of FIG. 2, including cameras, LiDAR sensors, radar sensors, and microphones. Further, the subject module, system, or function can include, for example, drive-by-wire system components (e.g., drive-by-wire system 202h of FIG. 2). Moreover, the subject module, system, or function can include, for example, autonomous vehicle compute (e.g., autonomous vehicle compute 400 of FIG. 4) components such as the perception system, planning system, database, localization system, and control system. In examples, machine learning models are used to process data associated with the module, system, or function. For ease of explanation, the present techniques are described as based on a subject modality. However, the error mining system can generate attributes and a dataset for any module, system, or function.

[0105] Accordingly, the configuration 604 enables the generation of a particular workflow related to the modules, systems, and / or functions to be evaluated using the generated datasets. In examples, the configuration 604 is specified in plain English. For example, the specified configuration describes a subject modality and the approach used to analyze the subject modality. Errors associated with the subject modality can be specified in the configuration, along with a modeling framework used to analyze the subject modality based on the generated datasets. Additionally, details associated withAttorney Docket No. 46154-0577WO1 / I2023201the modeling framework can be included in the configuration, such as a target modality used in cross modality training.

[0106] For example, consider a camera-based object detection model being trained to detect objects of various sizes. The specified configuration includes, for example, the an identification of the system to be trained as the camera system. The specified configuration can further include object detection as a functionality associated with the specified system. The modeling framework is also identified in the configuration. In examples, the modeling framework includes cross modality training, temporal training, and prior training.

[0107] Cross modality training leverages data from multiple, distinct modules, systems, or functionalities to train or refine a subject modality. By training models on data from multiple modalities, the trained system learns to compensate for the limitations the respective system. For instance, LiDARs can detect objects in poor visibility conditions where cameras might fail.

[0108] Temporal based training uses time-series data to train models based on patterns that occur over time. A model is trained to output accurate data where the temporal order or consistency of the output data is significant. For example, a camera system can experience “flickering,” where the camera-based object detection system switches between different classified detections over time for a same object.

[0109] Prior based training uses pre-existing knowledge or data (e g., priors) to train machine learning models. For example, prior maps or map queries are used to train models used for tasks like navigation, localization, and mapping.

[0110] By identifying the modules, systems, and / or errors to be evaluated using the generated datasets and a modeling framework, the attribute builder can use the specified configuration 604 to generate at least one attribute by the attribute generator. In examples, when the modeling framework is identified as cross modality training, a target modality is identified in the configuration 604. For example, when training a camera-based object detection model, a LiDAR-based object detection model can be identified as the target modality. The attribute generator 606 generates attributes that extract camera and LiDAR based object detection data from a database formatted intoAttorney Docket No. 46154-0577WO1 / I2023201parquet tables 608. In examples, the parquet tables 608 include driving data such as incoming raw drivelogs, auto-labeled logs, and the like.

[0111] The attribute generator 606 executes data mining on data in the parquet tables 608 according to the generated attributes. The output of the attribute generator is processed per agent records 610. The processed per agent records 610 are records identified according to the generated attributes. In examples, per-agent records extracted from parquet tables typically include detailed information about each individual agent (e.g., objects 104a-104n of FIG. 1) detected in the driving data. In examples the per-agent records 610 include a unique identifier for each agent, and a timestamp that identifies the specific time at which the data for respective agents is recorded. Additionally, in examples, the per-agent records 610 include spatial coordinates (e.g., latitude, longitude, altitude) and orientation (e.g., heading, pitch, roll) of the agent. The per-agent records 610 can also include information about the agent's speed and acceleration, and the dimensions and location of a bounding box encapsulating the agent, often used in object detection and tracking. The per-agent records 610 can include a type of agent (e.g., car, pedestrian, cyclist), which is used in classification tasks. Moreover, per-agent records 610 include raw or processed data from various sensors (e.g., camera images, LiDAR point clouds) associated with the agent. The per-agent records 610 can also include agent attributes such as color, make and model of vehicles, or specific behaviors (e.g., jaywalking for pedestrians). In some embodiments, auto-labeled annotations may be included that provide context or corrections to the detected data.

[0112] The processed per-agent records 610 are stored at the cloud storage 612. At the cloud storage 612, the per-agent records 610 are accessed to generate a dataset by the dataset generator 616. In some embodiments, mining queries 614 are used to refine the records stored in the cloud storage 612. Mining queries 614 can be generated by a user in real time to refine the per-agent records 610. For example, the records in cloud storage 612 are further refined by analytics 618. The analytics 618 execute the mining queries 614 at the cloud storage 612. The analytics 618 can include for example, an interactive query service used to analyze data using a programming language, such as structured query language (SQL). The analytics 618 can also visualize the records in cloud storage 612. For example, consider records in cloud storage 612 extracted toAttorney Docket No. 46154-0577WO1 / I2023201analyze dimensions associated with camera-based object detection. Mining queries 614 are used to apply a threshold of dimensions to bounding boxes with a bounding box size differential. Accordingly, analytics 618 enable an iterative refinement of the records in the cloud storage 612 based on identified patterns. In some embodiments, the thresholds are iteratively adjusted to obtain records including data within identified thresholds. Accordingly, analytics 618 enable distillation of records in cloud storage to obtain relevant data.

[0113] The dataset generator 620 builds a dataset 622 based on the refined records in the cloud storage 612. In examples, the dataset generator 620 aggregates the per-agent records. For example, the dataset generator 620 aggregates information included in the per-agent records such as position, velocity, and bounding boxes for respective agents overtime. In some embodiments, the records are labeled with agent attributes such as the type of agent (e.g., car, pedestrian), actions (e.g., stopping, turning), and interactions with other agents. Additionally, in some embodiments, the records are organized into temporal sequences to capture the movement and behavior of agents over time. In some embodiments, specific driving scenarios are generated by combining records that meet predetermined criteria (e.g., urban driving, highway driving). At block 622, the generated dataset is released for use in the model framework at block 624. In some embodiments, the released dataset is in a custom data format. The model frameworks are used at block 624 for evaluation, validation, model training, and model fine-tuning of a subject module, system, or modality. For example, the datasets generated from per-agent records are used to evaluate and test the performance of autonomous driving models. Metrics such as accuracy, precision, and recall are calculated based on the model’s ability to correctly detect and track agents.

[0114] In some embodiments, computing systems and machine learning models are deployed in dynamic environments where errors occur. These errors can stem from various sources, such as fluctuating sensor readings, inconsistent environmental conditions, or unexpected data inputs. In examples, sensor data exhibits anomalies due to temperature variations or electromagnetic interference. Additionally, in examples environmental data is inconsistent due to changes in weather or physical obstructions. FIGs. 8-14 show examples of errors that can occur during vehicle operation, such asAttorney Docket No. 46154-0577WO1 / I2023201autonomous vehicle operation. In some embodiment, the vehicle is the same as or similar to vehicle 200 of FIG. 2. These errors are detected and categorized according to their attributes based on, at least in part, a subject modality, modeling framework, or any combinations thereof. A dataset is generated based on records extracted from obtained data according to the attributes. The root causes of the detected errors are then analyzed using statistical techniques, machine learning, manual review, or any combinations thereof. Solutions are subsequently implemented to rectify the identified errors, ensuring the robustness and reliability of the computing systems and machine learning models in these challenging environments.

[0115] FIG. 7 shows a false positive in LIDAR based object detection and segmentation. In the example of FIG. 7, LiDAR based object detection is executed based on data associated with image 702. The LiDAR based object detection results in false positive detections due to water splashes on the obscuring data capture at the LIDAR. The image 702 shows no vehicle immediately behind the AV. However, the LiDAR based object detection outputs a detected object 706 in the mapping 704. Additionally, LiDAR based object detection is executed based on data associated with image 722. The image 722 shows no vehicle immediately behind the AV. However, the LiDAR based object detection outputs a segmented object 726 in the mapping 724.

[0116] In some embodiments, the false positive LiDAR based object detection and segmentation shown in FIG. 7 can be used in cross modality training with image based object detection. As shown in the images 702 and 722, no vehicles are detected near the AV. The configuration for input into an attribute generator will specify the source modality (e.g., LiDAR), the approach used (cross modality training), and target modality (e.g., camera). The attribute generator can generate attributes used to extract agent records where false positive object detection occurs in the LiDAR data when compared to the camera data.

[0117] FIG. 8 shows a false negative in camera based object detection. In the example of FIG. 8, a trailer in the image 804 is not detected by the camera based object detection. The missed detection is due to the location of the object at the camera border. In the corresponding mapping 802, the trailer is not detected by the camera based object detection model (ISN) corresponding to the deeper green line output. However, the LiDARAttorney Docket No. 46154-0577WO1 / I2023201based object detection model detects the trailer within the circle 806. The configuration for input into an attribute generator will specify the source modality (e.g., camera), the approach used (cross modality training), and target modality (e.g., LiDAR). The attribute generator can generate attributes used to extract agent records where false negative object detection occurs in the camera data when compared to the LiDAR data.

[0118] FIG. 9 shows a false positive in camera based object detection. In the example of FIG. 9, the camera based object detection system identifies a pedestrian 906 in the mapping 902. However, a pedestrian is not present in the corresponding image 904.

[0119] FIG. 10 shows a false negative in LiDAR based object detection. In the example of FIG. 10, a squatting construction worker in the image 1004 is not detected by the LiDAR based object detection at 1006. The missed detection is due to the position of the construction worker. In the corresponding mapping 1002, the construction worker is not detected by the LiDAR based object detection model.

[0120] FIG. 11 shows a false positive vehicle detection by a LiDAR-based object detection model. In the example of FIG. 11, a vehicle is detected in the mapping 1104 as being behind a construction zone sign in the real-world by the LiDAR-based object detection model. However, the detected vehicle does not exist in the image 1102.

[0121] FIGs. 12A and 12B show flickering predictions from a semantic model. As shown in the mappings 1204A and 1204B, the model type predicted by the sematic model flickers between multiple states that represent different object types. In the example of FIGs. 12A and 12B, the model types correspond to two different colors; red represents an emergency vehicle, and green represents a vehicle. The flickering outputs in the mappings 1204A and 1204B occur in the same scene, just milliseconds away in time.

[0122] FIG. 13 shows inaccurate predictions from a prediction model. In the example of FIG. 13, mapping 1302 captures the environment at timestamp TO, just a few milliseconds ahead of the mapping 1304 at timestamp T1. The prediction model predicts a straight trajectory of a vehicle at timestamp TO as shown in mapping 1302. However, at timestamp T1 the vehicle is shown as attempting a left turn in mapping 1304.

[0123] FIG. 14 shows inaccurate heading / flickering by a LiDAR-based object detection model. In the example of FIG. 14, a large construction vehicle is detected in theAttorney Docket No. 46154-0577WO1 / I2023201mapping 1404 as diagonally positioned across the roadway by the LiDAR-based object detection model. However, the large construction vehicle is positioned with a heading along the roadway as shown in image 1402.

[0124] Accordingly, the error mining system is highly scalable and adaptable to any raw, incoming datasets, significantly boosting recall and increasing the likelihood of obtaining larger batches of valuable data. The error mining system performs error discovery first, annotating data clips determined to be relevant to identified issues. This approach ensures that nearly the entire annotation effort directly contributes to model improvements, maximizing its value. The error mining system is capable of proactively discovering potential errors. The error mining system is highly modular, multipurpose, and configurable, enabling end-users to easily adapt the error mining system (via a configuration specified in plain English) to mine different error cases for multiple different autonomy modules without necessitating any code changes. The error mining system is well-designed and extensible, ensuring users the ability to easily develop new capabilities on top of the existing ones if needs emerge.

[0125] Referring now to FIG. 15, illustrated is a flowchart of a process 1500 that enables an error mining system. In some embodiments, one or more of the steps described with respect to process 1500 are performed (e.g., completely, partially, and / or the like) by an autonomous system 202 of FIG. 2. Additionally, or alternatively, in some embodiments one or more steps described with respect to process 1500 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.

[0126] At block 1502, attributes are generated based on a subject modality (e.g., model or function to be evaluated / validated / trained / finetuned) and modeling framework (e.g., ground truth free approaches including cross-model evaluation, temporal evaluation, and prior evaluation). In some embodiments, generating attributes includes selecting agents of interest based on pre-filtering or cross-modality evaluation results of the subject modality.

[0127] At block 1504, records are extracted from driving data (e.g., Parquet-based data, AV driving data, human driving data) according to the generated attributes. In some embodiments, the records are refined prior to generating the dataset based on patternsAttorney Docket No. 46154-0577WO1 / I2023201observed in the extracted records. Additionally or alternatively, the records are refined by applying thresholds derived from patterns observed in the extracted records.

[0128] At block 1506, a dataset is generated using the extracted records by at least aggregating the records into a dataset with a predetermined format (e.g., scene MMDB, custom-made parquet DBs, etc.). In some embodiments, aggregating the records into a predetermined format comprises auto-labeling the records or manually (e.g., by a human) annotating the records.

[0129] At block 1508, the subject modality is evaluated (e.g., evaluating, validating, training, finetuning) within the modeling framework. In some embodiments, the modeling framework includes cross-modality training (e.g., training distinct modalities), temporalbased training (e.g., flickering or displacement errors with trajectories), prior-based training (e.g., map queries), or any combinations thereof.

[0130] The described error mining includes extracting records from driving data and generating a dataset according to attributes. Performing error discovery before annotation means identifying potential errors in the data early on. This proactive approach enables annotations to be focused on datasets tailored to correct these errors, rather than annotating data blindly. By targeting known issues, the annotation process becomes more efficient and effective, directly contributing to model improvement. When annotation is focused on identified errors, it ensures that the time and resources spent on annotation are used optimally. Instead of annotating large volumes of data without direction, annotation is concentrated on areas associated with the most significant impact on model performance. This targeted approach enhances the quality of the training data, leading to better model accuracy and reliability.

[0131] The error mining is scalable, meaning it can be applied to large and diverse datasets without requiring extensive manual intervention. This scalability enables management and processing of large volumes of raw datasets, which often contain unstructured and noisy data. By systematically identifying and addressing errors, the present techniques improve data quality across various types of datasets. Moreover, the present techniques increase recall, which is the ability of the model to identify all relevant instances in the data. By proactively finding potential errors, models are trained on high- quality data, reducing the likelihood of excluding important information from model trainingAttorney Docket No. 46154-0577WO1 / I2023201processes. This proactive approach builds robust models that perform well in real-world scenarios.

[0132] Additionally, the present techniques enable effective solutions to long-tail problems. Long-tail problems refer to issues that occur infrequently but can have significant consequences, such as rare events in autonomous driving. The present techniques address these long-tail problems by identifying errors that might otherwise be overlooked, and using these errors for further training. By improving the model's ability to handle rare and complex scenarios associated with identified errors, the present techniques enable safer and more reliable autonomous systems.

[0133] Embodiments

[0134] According to some non-limiting embodiments or examples, provided is a device, including: a processor and memory configured to execute a method, including: generating attributes based on a subject modality and modeling framework; extracting records from driving data according to the generated attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; and evaluating the subject modality within the modeling framework.

[0135] According to some non-limiting embodiments or examples, provided is a system including: at least one computer-readable medium storing computer-executable instructions; at least one processor configured to execute the computer-executable instructions, the execution carrying out a method including generating attributes based on a subject modality and modeling framework; extracting records from driving data according to the generated attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; and evaluating the subject modality within the modeling framework.

[0136] According to some non-limiting embodiments or examples, provided is non-transitory computer-readable storage medium including at least one program for execution by at least one processor of a vehicle, the at least one program including instructions which, when executed by the at least one processor, cause the vehicle to perform the method including generating attributes based on a subject modality and modeling framework; extracting records from driving data according to the generatedAttorney Docket No. 46154-0577WO1 / I2023201attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; and evaluating the subject modality within the modeling framework

[0137] Further non-limiting aspects or embodiments are set forth in the following numbered embodiments:

[0138] Embodiment 1: A method, including: generating attributes based on a subject modality and modeling framework; extracting records from driving data according to the generated attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; and evaluating the subject modality within the modeling framework.

[0139] Embodiment 2: The method of any preceding embodiment, where the records are refined prior to generating the dataset based on patterns observed in the extracted records.

[0140] Embodiment 3: The method of any preceding embodiment, where the records are refined prior to generating the dataset by applying thresholds derived from patterns observed in the extracted records.

[0141] Embodiment 4: The method of any preceding embodiment, where aggregating the records into a predetermined format includes combining the records into intervals.

[0142] Embodiment 5: The method of any preceding embodiment, where aggregating the records into a predetermined format includes auto-labeling the records or manually annotating the records.

[0143] Embodiment 6: The method of any preceding embodiment, where generating attributes further includes selecting agents of interest based on pre-filtering or cross-modality evaluation results of the subject modality.

[0144] Embodiment 7: The method of any preceding embodiment, where the modeling framework includes cross-modality training, temporal-based training, prior-based training, or any combinations thereof.

[0145] Embodiment 8: A system including: at least one computer-readable medium storing computer-executable instructions; at least one processor configured to executeAttorney Docket No. 46154-0577WO1 / I2023201the computer-executable instructions, the execution carrying out the method of any one of embodiments 1-7.

[0146] Embodiment 9: A non-transitory computer-readable storage medium including at least one program for execution by at least one processor of a vehicle, the at least one program including instructions which, when executed by the at least one processor, cause the vehicle to perform the method of any one of embodiments 1-7.

[0147] 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

Attorney Docket No. 46154-0577WO1 / I2023201WHAT IS CLAIMED IS:

1. A method, comprising:generating attributes based on a subject modality and modeling framework;extracting records from driving data according to the generated attributes; generating a dataset using the extracted records by at least aggregating the records into a dataset with a predetermined format; andevaluating the subject modality within the modeling framework.

2. The method of claim 1, wherein the records are refined prior to generating the dataset based on patterns observed in the extracted records.

3. The method of claim 1, wherein the records are refined prior to generating the dataset by applying thresholds derived from patterns observed in the extracted records.

4. The method of claim 1, wherein aggregating the records into a predetermined format comprises combining the records into intervals.

5. The method of claim 1, wherein aggregating the records into a predetermined format comprises auto-labeling the records or manually annotating the records.

6. The method of claim 1, wherein generating attributes further comprises selecting agents of interest based on pre-filtering or cross-modality evaluation results of the subject modality.

7. The method of claim 1, wherein the modeling framework comprises crossmodality training, temporal-based training, prior-based training, or any combinations thereof.Attorney Docket No. 46154-0577WO1 / I20232018. A system comprising:at least one computer-readable medium storing computer-executable instructions; at least one processor configured to execute the computer-executable instructions, the execution carrying out the method of any one of claims 1-7.9 A non-transitory computer-readable storage medium comprising at least one program for execution by at least one processor of a vehicle, the at least one program including instructions which, when executed by the at least one processor, cause the vehicle to perform the method of any one of claims 1-7.