Systems and methods for sensor calibration

Abstract

Systems and methods for verifying sensor calibration are described. To verify calibration of a sensor system having several types of sensors, an object can be configured to have a substantially reflective portion so that the sensors can isolate the substantially reflective portion and sensor data from the sensors can be compared to determine whether the position of the substantially reflective portion detected by each sensor aligns. To calibrate a sensor system, each sensor can use and detect an object having known calibration features and the detected data can be compared to known calibration data associated with the object to determine whether each sensor is properly calibrated.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Application No. 62 / 777,637, filed October 10, 2018. Technical Field

[0003] The present invention generally relates to verifying sensor calibration. In particular, the present description relates to systems and methods for providing automated verification of sensor calibration. Background Art

[0004] Configuring vehicles so that they have autonomous navigation capabilities is becoming increasingly popular. For example, drones and self-driving cars can be configured to navigate autonomously throughout an environment. These vehicles can rely on sensors such as light detection and ranging sensors, RADAR (radar), and vision-based sensors to assist these vehicles in navigating within an environment. Many of these autonomous vehicles use multiple sensor types simultaneously. In addition, the positions of these sensors when they are mounted on the autonomous vehicle may change slightly due to, for example, vibrations of the vehicle when the vehicle is crossing a road. This may have a negative impact on the calibration of the sensors. Summary of the Invention

[0005] In at least one aspect of the disclosure, a system for verifying sensor calibration is provided. The system includes at least one object configured to have a substantially reflective portion. The system includes at least one light detection and ranging sensor configured to detect a distance to at least one location associated with the substantially reflective portion of the at least one object. The system includes at least one camera sensor configured to detect a light intensity value associated with the at least one location of the substantially reflective portion of the at least one object. The system includes a computer readable medium storing computer executable instructions; and at least one processor communicatively coupled to the at least one light detection and ranging sensor and the at least one camera sensor and configured to execute the computer executable instructions stored on the computer readable medium. When the at least one processor executes the computer executable instructions stored on the computer readable medium, the at least one processor performs operations to: receive sensor data associated with the detected distance to the at least one location; generate a first predicted aggregate location associated with the substantially reflective portion based on the sensor data associated with the detected distance to the at least one location; receive sensor data associated with the detected intensity value of the at least one location; generate a second predicted aggregate location associated with the substantially reflective portion based on the sensor data associated with the detected intensity value of the at least one location; and determine an alignment error value based on the first predicted aggregate location and the second predicted aggregate location.

[0006] The computer executable instructions can include a first algorithm for generating the first predicted aggregate location and a second algorithm for generating the second predicted aggregate location. The second algorithm can be different than the first algorithm. At least one of the first algorithm and the second algorithm can be modified if the alignment error value is greater than a first alignment error threshold. At least one of the first algorithm and the second algorithm can be modified such that the generated first predicted aggregate location and the generated second predicted aggregate location are substantially aligned. When the at least one processor executes the instructions, the at least one processor can perform operations to initiate a calibration process of the at least one camera sensor or the at least one light detection and ranging sensor if the alignment error value is greater than a second alignment error threshold.

[0007] The substantially non-reflective portion can include a black face. The substantially reflective portion can include a substantially white face.

[0008] In another aspect of the present invention, a method for verifying sensor calibration is provided. The method includes configuring at least one object to have a substantially reflective portion. The method includes detecting, using a first sensor, a distance to at least one location associated with the substantially reflective portion of the at least one object. The method includes detecting, using a second sensor, light intensity values ​​at one or more locations associated with the substantially reflective portion of the at least one object. The method includes receiving sensor data associated with the detected distance to the at least one location. The method includes generating, based on the sensor data associated with the detected distance to the at least one location, a first predicted aggregate position associated with the substantially reflective portion. The method includes receiving sensor data associated with the intensity value of the at least one detected location. The method includes generating, based on the sensor data associated with the intensity value of the at least one detected location, a second predicted aggregate position associated with the substantially reflective portion; and determining an alignment error value based on the first predicted aggregate position and the second predicted aggregate position.

[0009] The first predicted aggregate position may be generated using a first algorithm, and the second predicted aggregate position may be generated using a second algorithm. The method may further include: modifying at least one of the first algorithm and the second algorithm if the alignment error value is greater than a first alignment error threshold. The method may further include: modifying at least one of the first algorithm and the second algorithm so that the generated first predicted aggregate position and the generated second predicted aggregate position are substantially aligned. The method may further include: calibrating at least one of the at least one imaging sensor and the at least one light detection and ranging sensor if the alignment error value is greater than a second alignment threshold.

[0010] In another aspect of the present invention, a system for validating a sensor is provided. The system includes: at least one object including at least one calibration feature; and a vehicle. The vehicle includes a plurality of sensors configured to detect the at least one calibration feature of the at least one object as the at least one object and the vehicle move relative to each other. Each of the plurality of sensors has a plurality of parameters, wherein the plurality of parameters include a plurality of internal parameters and a plurality of external parameters. The vehicle includes: a computer-readable medium storing computer-executable instructions; and at least one processor configured to be communicatively coupled to the plurality of sensors and to execute the instructions stored on the computer-readable medium. When the at least one processor executes the instructions, the at least one processor performs operations to: receive known calibration feature data associated with the at least one calibration feature; receive detection feature data associated with the at least one calibration feature from each of the multiple sensors; compare the received known feature data with the received detection feature data; generate a calibration error value for each of the multiple sensors based on the comparison of the received known feature data with the received detection feature data; and if the calibration error value corresponding to at least one sensor of the multiple sensors is greater than a first calibration error threshold, determine that the at least one sensor is incorrectly calibrated.

[0011] When the at least one processor executes the instructions, the at least one processor may further perform operations to calculate a correction parameter for at least one internal parameter associated with the at least one sensor determined to be incorrectly calibrated. Calculating the correction parameter may be based, at least in part, on determining that the at least one sensor is incorrectly calibrated. When the at least one processor executes the instructions, the at least one processor may further perform operations to modify the at least one internal parameter based on the correction parameter.

[0012] Each sensor of the plurality of sensors may have at least one internal parameter. When the at least one processor executes the instructions, the at least one processor may further perform operations to receive internal data corresponding to the at least one internal parameter and calculate a correction parameter for the at least one internal parameter associated with the at least one sensor determined to be incorrectly calibrated. Calculating the correction parameter for the at least one internal parameter is based at least in part on the received internal data. The internal parameter may include at least one of: operating frequency, field of view, beam width, beam power, and signal-to-noise ratio.

[0013] When the at least one processor executes the instructions, the at least one processor can further perform operations to determine that the at least one sensor is in a fault state if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a second calibration error threshold. When the at least one processor executes the instructions, the at least one processor can perform operations to calculate a correction parameter for at least one extrinsic parameter associated with the at least one sensor determined to be in a fault state. The calculation of the correction parameter for the at least one extrinsic parameter can be based at least in part on the determination that the at least one sensor is in a fault state. The at least one extrinsic parameter can include at least one of a position of a sensor when mounted on the vehicle, a distance of a sensor relative to other sensors, an angle of a sensor, a noise level caused by an environment, and an ambient brightness of the environment.

[0014] The at least one object can include three faces defining three mutually perpendicular planes, each of the three faces including at least one calibration feature. The at least one object can include a fixed road feature.

[0015] The plurality of sensors can include a first sensor of a first type and a second sensor of a second type. The plurality of sensors can include at least one RADAR sensor, and the at least one object can include an internal metal core detectable by the at least one RADAR sensor. The plurality of sensors can include at least one light detection and ranging sensor and at least one RADAR sensor, and the at least one object can include an augmented portion. The augmented portion is substantially detectable by the at least one light detection and ranging sensor and substantially undetectable by the at least one RADAR sensor.

[0016] The known feature data can include at least one of information received from a remote sensor, information received from one or more sensors at a previous time, and information determined by a pre-calibrated sensor of the one or more sensors. The known feature data can include information received from a remote sensor mounted on a second vehicle. The known feature data can include information received from a sensor of a first type, and the plurality of sensors can include at least one sensor of a second type different from the first type.

[0017] When the at least one processor executes the instructions, the at least one processor can perform operations to cause the vehicle to stop operating if the calibration error value corresponding to the at least one sensor determined to be improperly calibrated is greater than a calibration error threshold. When the at least one processor executes the instructions, the at least one processor can perform operations to notify a remote technician if the calibration error value corresponding to the at least one sensor determined to be improperly calibrated is greater than a calibration error threshold. When the at least one processor executes the instructions, the at least one processor performs operations to navigate the vehicle to a service location if the calibration error value corresponding to the at least one sensor determined to be improperly calibrated is greater than a calibration error threshold. When the at least one processor executes the instructions, the at least one processor can perform operations to disable the at least one sensor if the calibration error value corresponding to the at least one sensor determined to be improperly calibrated is greater than a calibration error threshold.

[0018] In another aspect of the disclosure, a method for calibrating sensors is provided. The method includes causing a vehicle and at least one object to move relative to each other, the vehicle including a plurality of sensors and the at least one object including at least one calibration feature. The method includes detecting, with the plurality of sensors, the at least one calibration feature of the at least one object as the at least one object and the vehicle are moving relative to each other, the plurality of sensors each having a plurality of parameters, the plurality of parameters including a plurality of internal parameters and a plurality of external parameters. The method includes receiving known calibration feature data associated with the at least one calibration feature. The method includes receiving, from each sensor of the plurality of sensors, detected feature data associated with the at least one calibration feature. The method includes comparing the received known feature data to the received detected feature data. The method includes generating, for each sensor of the plurality of sensors, a calibration error value based on the comparison of the received known feature data to the received detected feature data. The method includes determining that at least one sensor of the plurality of sensors is improperly calibrated if the calibration error value corresponding to the at least one sensor is greater than a first calibration error threshold.

[0019] The method can further include calculating a correction parameter for at least one internal parameter associated with the at least one sensor determined to be improperly calibrated. Calculating the correction parameter can be based at least in part on determining that the at least one sensor is improperly calibrated. The method can further include modifying the at least one internal parameter based on the correction parameter.

[0020] Each of the plurality of sensors may include one or more monitoring devices configured to detect internal data corresponding to at least one internal parameter. The method may further include: receiving internal data associated with the at least one internal parameter; and calculating a correction parameter for the at least one internal parameter of the at least one sensor determined to be incorrectly calibrated based at least in part on the received internal data. The method may further include: determining that the at least one sensor is in a fault state if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a second calibration error threshold. The method may further include: calculating a correction parameter for at least one external parameter associated with the at least one sensor determined to be incorrectly calibrated. The calculation of the correction parameter for the at least one external parameter may be based at least in part on determining that the at least one sensor is in a fault state.

[0021] The at least one object may comprise three faces defining three mutually perpendicular planes, each of the three faces comprising at least one calibration feature.The at least one object may comprise a fixed road feature.

[0022] The plurality of sensors may include at least one RADAR sensor, and the at least one object may include an internal metal core detectable by the at least one RADAR sensor. The plurality of sensors may include at least one light detection and ranging sensor and at least one RADAR sensor. The at least one object may include an enhanced portion that is substantially detectable by the at least one light detection and ranging sensor and substantially undetectable by the at least one RADAR sensor. The known characteristic data may include information received from a remote sensor, information received at a previous time from one or more sensors, information determined by a pre-calibrated sensor of the one or more sensors. The known characteristic data may include information received from a remote sensor mounted on a second vehicle. The known characteristic data may include information received from a first type of sensor, and the plurality of sensors may include at least one sensor of a second type that is different from the first type.

[0023] The method may include causing the vehicle to cease operation if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold. The method may include notifying a remote technician if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold. The method may include navigating the vehicle to a service location if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold. The method may include disabling the at least one sensor if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold.

[0024] These and other aspects, features, and implementations may be expressed as methods, apparatus, systems, components, program products, methods or steps for performing the functions, and other means.

[0025] These and other aspects, features and implementations will be apparent from the following description including the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Illustrate an example of an autonomous vehicle with autonomous capabilities.

[0027] Figure 2 An example "cloud" computing environment is illustrated.

[0028] Figure 3 An example computer system.

[0029] Figure 4 Illustrate an example architecture for an autonomous vehicle.

[0030] Figure 5 Illustrate examples of inputs and outputs that can be used by the perception module.

[0031] Figure 6 An example of a LiDAR system is illustrated.

[0032] Figure 7 Illustrate a LiDAR system in operation.

[0033] Figure 8 Additional details of the operation of the LiDAR system are illustrated.

[0034] Figure 9 A block diagram illustrating the relationship between the inputs and outputs of the planning module.

[0035] Figure 10 Illustrate a directed graph used in path planning.

[0036] Figure 11A block diagram illustrating the inputs and outputs of the control module.

[0037] Figure 12 A block diagram illustrating the controller's inputs, outputs, and components.

[0038] Figure 13 is a diagram illustrating a system for verifying sensor calibration according to one or more embodiments of the present invention.

[0039] Figure 14 is an illustrative example of generating a first predicted aggregate position and a second predicted aggregate position according to one or more embodiments of the present invention.

[0040] Figure 15 is a flow chart describing a method for verifying sensor calibration according to one or more embodiments of the present invention.

[0041] Figure 16 is a diagram illustrating a system for calibrating a sensor according to one or more embodiments of the present invention.

[0042] Figure 17 is a flow chart describing a method for calibrating a sensor according to one or more embodiments of the present invention.

[0043] Figure 18 is a flow chart describing a method for modifying internal parameters based on calibration errors according to one or more embodiments of the present invention.

[0044] Figure 19 is a flow chart describing a method for determining sensor failure based on calibration error according to one or more embodiments of the present invention. DETAILED DESCRIPTION

[0045] In the following description, for the purpose of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, it will be apparent that the present invention can be practiced without these specific details. In other examples, well-known configurations and devices are shown in block diagram form to avoid unnecessarily obscuring the present invention.

[0046] In the accompanying drawings, for ease of description, a specific arrangement or order of schematic elements (such as those representing devices, modules, instruction blocks, and data elements) is shown. However, it will be understood by those skilled in the art that the specific order or arrangement of schematic elements in the accompanying drawings is not intended to require a specific processing order or sequence, or separation of processing processes. In addition, the inclusion of schematic elements in the accompanying drawings is not intended to mean that such elements are required in all embodiments, nor is it intended to mean that the features represented by such elements cannot be included in some embodiments or cannot be combined with other elements in some embodiments.

[0047] Furthermore, in the drawings, connecting elements, such as a line or arrow or both, between two or more other depicted components indicate either a connection, an association, or a relationship between the other depicted components. Absence of such connecting elements between components, however, is not intended to imply that a connection between the components does not exist, or that the components do not affect each other. In other words, some connections, associations, or relationships between components are not shown in the drawings so as not to obscure the disclosure. Moreover, use of a single connecting element to represent multiple connections, associations, or relationships between components is not intended to limit the scope of the application to only such explicitly illustrated connections, associations, or relationships.

[0048] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings. In the following detailed description of embodiments, 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 skilled in the art that the various described embodiments can be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0049] Several of the described features can each be used alone or in combination with any of the other features. However, none of the individual features maybe sufficient to solve any of the problems discussed above, or to address any of the other concerns described herein. Some of the problems discussed above can not be solved by any of the features described herein. Although the following claims list certain combinations, other combinations are also possible and are contemplated. Although provided with headings, information relating to a particular heading can also be found elsewhere in the specification. Embodiments are described herein in accordance with the following outline:

[0050] 1. OVERALL SUMMARY

[0051] 2. SYSTEM SUMMARY

[0052] 3. AUTONOMOUS VEHICLE ARCHITECTURE

[0053] 4. AUTONOMOUS VEHICLE INPUT

[0054] 5. AUTONOMOUS VEHICLE PLANNING

[0055] 6. AUTONOMOUS VEHICLE CONTROL

[0056] 7. SYSTEM AND METHOD FOR VERIFICATION OF SENSOR CALIBRATION

[0057] 8. SYSTEM AND METHOD FOR VERIFICATION OF SENSOR CALIBRATION

[0058] OVERALL SUMMARY

[0059] This disclosure describes, among other things, techniques for automated verification of sensor calibration. Automated verification of sensor calibration can reduce the time required to ensure an autonomous vehicle's sensors are accurate to facilitate safe navigation. Systems and methods exploit the inherent properties of various sensor types to provide efficient and accurate techniques for calibrating each sensor itself.

[0060] Vehicles (e.g., drones, self-driving cars, etc.) can be configured to navigate autonomously throughout an environment. These vehicles can rely on sensors such as light detection and ranging sensors, RADAR, and vision-based sensors to assist these vehicles in navigating within an environment. Many of these autonomous vehicles use multiple sensor types simultaneously. In addition, the positions of these sensors when they are mounted on the autonomous vehicle may change slightly due to, for example, the vibration of the vehicle when the vehicle is crossing a road. This may have a negative impact on the calibration of the sensors.

[0061] Because these vehicles use multiple sensor types simultaneously, a calibration process may be required to combine data from different sensors into a common reference frame. Traditional calibration methods may require manual measurements and inputs. These methods can be laborious and time-consuming, and require the vehicle to be out of service for extended periods of time. Therefore, it may be desirable to provide techniques for enabling vehicles to self-calibrate their sensors in a timely and efficient manner. It may also be desirable to provide techniques for enabling vehicles to calibrate their sensors while navigating an environment.

[0062] System Overview

[0063] Figure 1 An example of an autonomous vehicle 100 having autonomous capabilities is illustrated.

[0064] As used herein, the term "autonomous capability" refers to a function, feature, or facility that enables a vehicle to operate partially or fully without real-time human intervention, including but not limited to fully autonomous vehicles, highly autonomous vehicles, and conditionally autonomous vehicles.

[0065] As used herein, an autonomous vehicle (AV) is a vehicle with autonomous capabilities.

[0066] As used herein, "vehicle" includes any mode of transport for goods or people, such as a car, bus, train, airplane, drone, truck, boat, ship, submersible, or spacecraft. An unmanned car is an example of a vehicle.

[0067] As used herein, a "trajectory" refers to a path or route that navigates an AV from a first spatiotemporal location to a second spatiotemporal location. In embodiments, the first spatiotemporal location is referred to as an initial location or a starting location, and the second spatiotemporal location is referred to as a destination, a final location, a target, a target location, or a target location. In some examples, a trajectory is composed of one or more segments (e.g., pieces of a road), and each segment is composed of one or more blocks (e.g., a portion of a lane or an intersection). In embodiments, a spatiotemporal location corresponds to a real-world location. For example, a spatiotemporal location is a pickup or drop-off location for a person or cargo.

[0068] As used herein, a "(one or more) sensor" includes one or more hardware components for detecting information related to the sensor's surrounding environment. Some hardware components can include sensing components (e.g., image sensors, biometric sensors), transmitting and / or receiving components (e.g., laser or radio frequency wave emitters and receivers), electronic components such as analog-to-digital converters, data storage devices such as RAM and / or non-volatile memory, software or firmware components, and data processing components such as application-specific integrated circuits, microprocessors, and / or microcontrollers.

[0069] As used herein, a "scene description" is a data structure (e.g., a list) or data stream that includes one or more classified or labeled objects detected by one or more sensors on an AV vehicle or provided by a source external to the AV.

[0070] As used herein, a "road" is a physical area that can be traversed by a vehicle, and can correspond to a named thoroughfare (e.g., a city street, an interstate highway, etc.) or can correspond to an unnamed thoroughfare (e.g., a driveway within a house or office building, a section of a parking lot, a section of an empty parking lot, a dirt path in a rural area, etc.). Since some vehicles (e.g., four-wheel drive pickup trucks, sport utility vehicles (SUVs), etc.) are capable of traversing a variety of physical areas that are not particularly well-suited for vehicle travel, a "road" can be any physical area that has not been formally defined as a thoroughfare by a municipality or other governmental or administrative body.

[0071] As used herein, a "lane" is a portion of a road that can be traversed by vehicles and may correspond to most or all of the space between lane markings, or only a portion of the space between lane markings (e.g., less than 50%). For example, a road with lane markings that are far apart may accommodate two or more vehicles, such that one vehicle can pass another without crossing the lane markings and thus may be interpreted as having a lane narrower than the space between the lane markings, or as having two lanes between the lanes. Lanes may also be interpreted in the absence of lane markings. For example, lanes may be defined based on physical features of the environment (e.g., rocks in a rural area and trees along an avenue).

[0072] “One or more” includes a function performed by one element, a function performed by multiple elements, such as in a distributed manner, several functions performed by one element, several functions performed by several elements, or any combination of the foregoing.

[0073] It will also be understood that although in some cases, the terms "first," "second," etc. are used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first contact may be referred to as a second contact, and similarly, a second contact may be referred to as a first contact, without departing from the scope of the various described embodiments. Both the first contact and the second contact are contacts, but they are not the same contact.

[0074] The terms used in the specification of the various embodiments described herein are for the purpose of describing specific embodiments only and are not intended to be limiting. As used in the specification of the various embodiments described and the appended claims, the singular forms "a", "an", and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that "and / or" as used herein refers to and includes any and all possible combinations of one or more related list items. It will also be understood that when the terms "comprises", "comprising", "having" and / or "having" are used in this specification, the presence of the stated features, integers, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof is not excluded.

[0075] As used herein, the term "if' is, optionally, understood to mean "when" or "when a' or "in response to a determination that" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" is, optionally, understood to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]," or "in response to detecting [the stated condition or event]," depending on the context.

[0076] As used herein, an AV system refers to an array of AVs and hardware, software, stored data, and real-time generated data that support operation of the AVs. In embodiments, the AV system is incorporated within the AVs. In embodiments, the AV system is distributed across several locations. For example, some software of the AV system is implemented on a cloud computing environment 200 described below with respect to Figure 2

[0077] In general, this document describes techniques applicable to any vehicle with one or more levels of autonomy, including fully autonomous vehicles, highly autonomous vehicles, and conditionally autonomous vehicles, such as so-called Level 5, Level 4, and Level 3 vehicles, respectively (see SAE International Standard J3016: Taxonomy and Definitions for Terms Related to Driving Automation Systems for Road Vehicles, incorporated by reference in its entirety for additional details on levels of vehicle autonomy). The techniques described in this document are also applicable to partially autonomous vehicles and driver-assist vehicles, such as so-called Level 2 and Level 1 vehicles (see SAE International Standard J3016: Taxonomy and Definitions for Terms Related to Driving Automation Systems for Road Vehicles). In embodiments, one or more Level 1, Level 2, Level 3, Level 4, and Level 5 vehicle systems can automatically perform certain vehicle operations (e.g., steering, braking, and use of a map) under certain operating conditions based on processing of sensor inputs. The techniques described in this document can benefit vehicles at any level, ranging from fully autonomous vehicles to human-operated vehicles.

[0078] Referring to Figure 1 , the AV system 120 causes the AV 100 to operate along a trajectory 198 through the environment 190 to a destination 199 (sometimes referred to as a final location) while avoiding objects (e.g., natural obstacles 191, vehicles 193, pedestrians 192, cyclists, and other obstacles) and obeying road rules (e.g., operating rules or driving preferences).

[0079] In embodiments, the AV system 120 includes the apparatus 101 for receiving and operating on operational commands from the computer processor 146. In embodiments, the computer processor 146 is in communication with the cloud computing environment 200 described below with respect to Figure 3 ​The described processor 304 is similar. Examples of the apparatus 101 include a steering controller 102, a brake 103, a gearshift, an accelerator pedal or other acceleration control mechanism, a windshield wiper, a side door lock, a window control, and a turn signal.

[0080] In embodiments, the AV system 120 includes sensors 121 for measuring or inferring properties of the state or condition of the AV 100, such as the AV's position, linear and angular velocities and linear and angular accelerations, and heading (e.g., the direction of the AV 100's front end). Examples of the sensors 121 are a GPS, an inertial measurement unit (IMU) that measures both vehicle linear acceleration and angular rate, wheel rate sensors for measuring or estimating wheel slip, wheel brake pressure or brake torque sensors, engine torque or wheel torque sensors, and steering angle and angular rate sensors.

[0081] In embodiments, the sensors 121 also include sensors for sensing or measuring properties of the AV's environment. For example, monocular or stereo video cameras 122 in the visible, infrared, or thermal (or both) light spectrum, LiDAR 123, RADAR, ultrasonic sensors, time-of-flight (TOF) depth sensors, speed sensors, temperature sensors, humidity sensors, and precipitation sensors.

[0082] In embodiments, the AV system 120 includes a data storage unit 142 and a memory 144 for storing machine instructions associated with a computer processor 146 or data collected by the sensors 121. In embodiments, the data storage unit 142 is similar to the ROM 308 or storage 310 described below with respect to the AV 100. Figure 3 The described ROM 308 or storage 310 is similar. In embodiments, the memory 144 is similar to the main memory 306 described below. In embodiments, the data storage unit 142 and the memory 144 store historical, real-time, and / or predictive information about the environment 190. In embodiments, the stored information includes maps, driving performance, traffic congestion updates, or weather conditions. In embodiments, data about the environment 190 is transmitted from the remote database 134 to the AV 100 over a communication channel.

[0083] In an embodiment, the AV system 120 includes communication devices 140 for transmitting measured or inferred attributes of the state and condition of other vehicles (such as position, linear and angular velocity, linear and angular acceleration, and linear and angular heading) to the AV 100. These devices include vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication devices, as well as devices for wireless communication via point-to-point or ad hoc networks, or both. In an embodiment, the communication devices 140 communicate across the electromagnetic spectrum (including radio and optical communications) or other media (e.g., air and acoustic media). The combination of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications (and in some embodiments, one or more other types of communications) is sometimes referred to as vehicle-to-everything (V2X) communication. V2X communications typically conform to one or more communication standards for communication with and between autonomous vehicles.

[0084] In an embodiment, the communication device 140 includes a communication interface. For example, a wired, wireless, WiMAX, Wi-Fi, Bluetooth, satellite, cellular, optical, near-field, infrared, or radio interface. The communication interface transmits data from the remote database 134 to the AV system 120. In an embodiment, the remote database 134 is embedded in a computer such as Figure 2 The communication interface 140 transmits data collected from the sensors 121 or other data related to the operation of the AV 100 to the remote database 134. In an embodiment, the communication interface 140 transmits information related to teleoperation to the AV 100. In some embodiments, the AV 100 communicates with other remote (e.g., "cloud") servers 136.

[0085] In an embodiment, the remote database 134 also stores and transmits digital data (e.g., data such as roads and street locations). This data is stored in the memory 144 on the AV 100 or transmitted from the remote database 134 to the AV 100 via a communication channel.

[0086] In an embodiment, the remote database 134 stores and transmits historical information regarding driving attributes (e.g., speed and acceleration profiles) of vehicles that have previously traveled along the trajectory 198 at similar times of day. In one implementation, such data may be stored in the memory 144 on the AV 100 or transmitted from the remote database 134 to the AV 100 via a communication channel.

[0087] The computing device 146 located on the AV 100 algorithmically generates control actions based on both real-time sensor data and a priori information, allowing the AV system 120 to perform its autonomous driving capabilities.

[0088] In an embodiment, the AV system 120 includes a computer peripheral device 132 coupled to a computing device 146 for providing information and alerts to a user of the AV 100 (e.g., an occupant or a remote user) and receiving input from the user. In an embodiment, the peripheral device 132 is similar to the one described below with reference to Figure 3 The display 312, input device 314 and cursor control 316 are discussed. The coupling may be wireless or wired. Any two or more of the interface devices may be integrated into a single device.

[0089] Figure 2 Illustrate an example "cloud" computing environment. Cloud computing is a service delivery model for enabling convenient, on-demand access to a shared pool of configurable computing resources (e.g., networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) over a network. In a typical cloud computing system, one or more large cloud data centers house the machines used to deliver the services provided by the cloud. Now refer to Figure 2 , cloud computing environment 200 includes cloud data centers 204a, 204b, and 204c interconnected by a cloud 202. Data centers 204a, 204b, and 204c provide cloud computing services to computer systems 206a, 206b, 206c, 206d, 206e, and 206f connected to cloud 202.

[0090] The cloud computing environment 200 includes one or more cloud data centers. Generally speaking, a cloud data center (e.g. Figure 2 The cloud data center 204a shown in FIG refers to a cloud (eg Figure 2 The physical arrangement of servers in a cloud 202 (or a specific portion of a cloud) as shown in FIG. For example, servers are physically arranged into rooms, groups, rows, and racks in a cloud data center. A cloud data center has one or more zones, which include one or more server rooms. Each room has one or more rows of servers, and each row includes one or more racks. Each rack includes one or more individual server nodes. In some implementations, servers in zones, rooms, racks, and / or rows are arranged into groups based on the physical infrastructure requirements of the data center facility, including power, energy, heat, heat sources, and / or other requirements. In an embodiment, the server nodes are similar to Figure 3 The data center 204a has many computing systems distributed across multiple racks.

[0091] Cloud 202 includes cloud data centers 204a, 204b, and 204c, as well as networks and network resources (e.g., network devices, nodes, routers, switches, and network cables) used to connect cloud data centers 204a, 204b, and 204c and facilitate access to cloud computing services by computing systems 206a-f. In embodiments, the network represents any combination of one or more local networks, wide area networks, or internetworks coupled via wired or wireless links deployed using terrestrial or satellite connections. Data exchanged over the network is transmitted using a variety of network layer protocols, such as Internet Protocol (IP), Multiprotocol Label Switching (MPLS), Asynchronous Transfer Mode (ATM), Frame Relay, and the like. Furthermore, in embodiments where the network represents a combination of multiple subnetworks, a different network layer protocol is used on each underlying subnetwork. In some embodiments, the network represents one or more interconnected internetworks, such as the public Internet.

[0092] Computing systems 206a-f, or cloud computing service consumers, are connected to the cloud 202 via network links and network adapters. In embodiments, computing systems 206a-f are implemented as various computing devices, such as servers, desktops, laptops, tablets, smartphones, Internet of Things (IoT) devices, autonomous vehicles (including cars, drones, shuttles, trains, buses, etc.), and consumer electronics. In embodiments, computing systems 206a-f are implemented in other systems or as part of other systems.

[0093] Figure 3 Illustrated is a computer system 300. In implementation, the computer system 300 is a special-purpose computing device. The special-purpose computing device is hard-wired to perform these techniques, or includes a digital electronic device such as one or more application-specific integrated circuits (ASICs) or field-programmable gate arrays (FPGAs) that are permanently programmed to perform the above-mentioned techniques, or may include one or more general-purpose hardware processors that are programmed to perform these techniques according to program instructions in firmware, memory, other memory, or a combination thereof. Such a special-purpose computing device may also combine customized hard-wired logic, ASICs, or FPGAs with customized programming to accomplish these techniques. In various embodiments, the special-purpose computing device is a desktop computer system, a portable computer system, a handheld device, a network device, or any other device that includes hard-wired and / or program logic to implement these techniques.

[0094] In an embodiment, computer system 300 includes a bus 302 or other communication mechanism for communicating information, and a hardware processor 304 coupled to bus 302 for processing information. Hardware processor 304 is, for example, a general-purpose microprocessor. Computer system 300 also includes a main memory 306, such as a random access memory (RAM) or other dynamic storage device, coupled to bus 302 to store information and instructions for execution by processor 304. In one implementation, main memory 306 is used to store temporary variables or other intermediate information during the execution of instructions to be executed by processor 304. When these instructions are stored in a non-transitory storage medium accessible to processor 304, computer system 300 becomes a special-purpose machine customized to perform the operations specified in the instructions.

[0095] In an embodiment, computer system 300 also includes a read-only memory (ROM) 308 or other static storage device coupled to bus 302 for storing static information and instructions for processor 304. A storage device 310, such as a magnetic disk, optical disk, solid-state drive, or three-dimensional cross-point memory, is provided and coupled to bus 302 to store information and instructions.

[0096] In an embodiment, the computer system 300 is coupled via a bus 302 to a display 312, such as a cathode ray tube (CRT), a liquid crystal display (LCD), a plasma display, a light emitting diode (LED) display, or an organic light emitting diode (OLED) display, for displaying information to a computer user. An input device 314, including alphanumeric and other keys, is coupled to the bus 302 for communicating information and command selections to the processor 304. Another type of user input device is a cursor controller 316, such as a mouse, a trackball, a touch-sensitive display, or cursor direction keys, for communicating direction information and command selections to the processor 304 and for controlling movement of a cursor on the display 312. Such input devices typically have two degrees of freedom along two axes, a first axis (e.g., an x-axis) and a second axis (e.g., a y-axis), which allow the device to specify a position on a plane.

[0097] According to one embodiment, the techniques herein are performed by computer system 300 in response to processor 304 executing one or more sequences of one or more instructions contained in main memory 306. These instructions are read into main memory 306 from another storage medium, such as storage device 310. Execution of the sequences of instructions contained in main memory 306 causes processor 304 to perform the process steps described herein. In alternative embodiments, hard-wired circuitry is used in place of or in combination with software instructions.

[0098] As used herein, the term "storage medium" refers to any non-transitory medium that stores data and / or instructions that cause a machine to operate in a particular manner. Such storage media include non-volatile media and / or volatile media. Non-volatile media include, for example, optical disks, magnetic disks, solid-state drives, or three-dimensional cross-point memory, such as storage device 310. Volatile media include dynamic memory, such as main memory 306. Common forms of storage media include, for example, floppy disks, diskettes, hard disks, solid-state drives, magnetic tape or any other magnetic data storage medium, CD-ROMs, any other optical data storage medium, any physical medium with a hole pattern, RAM, PROM, EPROM, FLASH-EPROM, NV-RAM, or any other memory chip or storage cartridge.

[0099] Storage media are distinct from transmission media, but can be used in conjunction with them. Transmission media participate in the transmission of information between storage media. Examples of transmission media include coaxial cables, copper wire, and optical fiber, including the wires that comprise bus 302. Transmission media can also take the form of acoustic or optical waves, such as those generated during radio wave and infrared data communications.

[0100] In one embodiment, various forms of media are involved in carrying one or more sequences of one or more instructions to processor 304 for execution. For example, the instructions may initially be executed on a disk or solid-state drive of a remote computer. The remote computer loads the instructions into its dynamic memory and sends the instructions over a telephone line using a modem. The modem local to computer system 300 receives the data on the telephone line and uses an infrared transmitter to convert the data to an infrared signal. An infrared detector receives the data carried in the infrared signal, and appropriate circuitry places the data on bus 302. Bus 302 carries the data to main memory 306, from which processor 304 retrieves and executes the instructions. The instructions received by main memory 306 may optionally be stored on storage device 310 before or after execution by processor 304.

[0101] Computer system 300 also includes a communication interface 318 coupled to bus 302. Communication interface 318 provides a two-way data communication coupled to a network link 320 that is connected to a local network 322. For example, communication interface 318 is an integrated services digital network (ISDN) card, a cable modem, a satellite modem, or a modem for providing a data communication connection with a corresponding type of telephone line. As another example, communication interface 318 is a local area network (LAN) card for providing a data communication connection with a compatible LAN. In some implementations, a wireless link is also implemented. In any such implementation, communication interface 318 sends and receives electrical, electromagnetic, or optical signals that carry digital data streams representing various types of information.

[0102] Network link 320 typically provides data communication to other data devices through one or more networks. For example, network link 320 provides a connection to a host computer 324 or to a cloud data center or facility operated by an Internet Service Provider (ISP) 326 through a local network 322. ISP 326, in turn, provides data communication services through the worldwide packet data communication network now commonly referred to as the "Internet" 328. Both local network 322 and Internet 328 use electrical, electromagnetic, or optical signals that carry digital data streams. The signals through the various networks and the signals on network link 320 and through communication interface 318 are example forms of transmission media, where these signals carry digital data to and from computer system 300. In an embodiment, network 320 includes cloud 202 or a portion of cloud 202 as described above.

[0103] Computer system 300 sends messages and receives data, including program code, through network(s), network link 320, and communication interface 318. In an embodiment, computer system 300 receives code for processing. The received code is executed by processor 304 upon receipt and / or stored in storage device 310 or other non-volatile storage for later execution.

[0104] Autonomous Vehicle Architecture

[0105] Figure 4 Examples for autonomous vehicles (e.g. Figure 1 100). The architecture 400 includes a sensing module 402 (sometimes referred to as sensing circuitry), a planning module 404 (sometimes referred to as planning circuitry), a control module 406 (sometimes referred to as control circuitry), a positioning module 408 (sometimes referred to as positioning circuitry), and a database module 410 (sometimes referred to as database circuitry). Each module plays a role in the operation of the AV 100. Collectively, the modules 402, 404, 406, 408, and 410 may be Figure 1 4 and 5. In some embodiments, any of modules 402, 404, 406, 408, and 410 is a combination of computer software (e.g., executable code stored on a computer-readable medium) and computer hardware (e.g., one or more microprocessors, microcontrollers, application-specific integrated circuits [ASICs], hardware memory devices, other types of integrated circuits, other types of computer hardware, or a combination of any or all of these).

[0106] In use, the planning module 404 receives data representing a destination 412 and determines data representing a trajectory 414 (sometimes referred to as a route) that the AV 100 may travel in order to reach (e.g., arrive at) the destination 412. In order for the planning module 404 to determine the data representing the trajectory 414, the planning module 404 receives data from the perception module 402, the positioning module 408, and the database module 410.

[0107] The perception module 402 uses, for example, Figure 1 One or more sensors 121 are shown to identify nearby physical objects, classify the objects (e.g., into types such as pedestrians, bicycles, cars, traffic signs, etc.), and provide a scene description including the classified objects 416 to the planning module 404.

[0108] The planning module 404 also receives data representing the AV's position 418 from the positioning module 408. The positioning module 408 determines the AV's position by using data from the sensor 121 and data from the database module 410 (e.g., geographic data) to calculate the position. For example, the positioning module 408 uses data from a GNSS (Global Navigation Satellite System) sensor and geographic data to calculate the longitude and latitude of the AV. In an embodiment, the data used by the positioning module 408 includes a high-precision map with lane geometry, a map describing the road network connectivity attributes, a map describing the physical attributes of the lanes (such as traffic speed, traffic volume, the number of vehicle and bicycle lanes, lane width, lane traffic direction, or lane marking type and location, or a combination thereof), and a map describing the spatial location of road features (such as intersections, traffic signs, or various types of other driving signals).

[0109] The control module 406 receives data representing the trajectory 414 and data representing the AV's position 418 and operates the AV's control functions 420 a - 420 c (e.g., steering, throttle, brakes, ignition) in a manner that will cause the AV 100 to travel the trajectory 414 to reach the destination 412. For example, if the trajectory 414 includes a left turn, the control module 406 will operate the control functions 420 a - 420 c in such a manner that the steering angle of the steering function will cause the AV 100 to turn left, and the throttle and brakes will cause the AV 100 to pause and wait for a passing pedestrian or vehicle before executing the turn.

[0110] Autonomous Vehicle Input

[0111] Figure 5 Example perception module 402 ( Figure 4 ) used by the inputs 502a-502d (e.g., Figure 11) and outputs 504a-504d (e.g., sensor data). One input 502a is a LiDAR (Light Detection and Ranging) system (e.g., Figure 1 LiDAR 123 is shown. LiDAR is a technology that uses light (e.g., a beam of light such as infrared light) to obtain data about physical objects in its line of sight. The LiDAR system generates LiDAR data as output 504a. For example, LiDAR data is a collection of 3D or 2D points (also called a point cloud) used to construct a representation of the environment 190.

[0112] Another input 502b is a RADAR (radar) system. RADAR is a technology that uses radio waves to obtain data about nearby physical objects. RADAR can obtain data about objects that are not within the line of sight of the LiDAR system. RADAR system 502b generates RADAR data as output 504b. For example, RADAR data is one or more radio frequency electromagnetic signals used to construct a representation of environment 190.

[0113] Another input 502c is a camera system. The camera system uses one or more cameras (e.g., a digital camera using a light sensor such as a charge coupled device [CCD]) to acquire information about nearby physical objects. The camera system produces camera data as output 504c. The camera data is typically in the form of image data (e.g., data in an image data format such as RAW, JPEG, PNG, etc.). In some examples, the camera system has multiple independent cameras, such as for the purpose of stereo imaging (stereo vision), which enables the camera system to perceive depth. Although the objects perceived by the camera system are described here as "nearby", this is relative to the AV. In use, the camera system can be configured to "see" objects that are far away (e.g., up to 1 km or more in front of the AV). Therefore, the camera system can have features such as sensors and lenses that are optimized for perceiving distant objects.

[0114] Another input 502d is a traffic light detection (TLD) system. The TLD system uses one or more cameras to obtain information about traffic lights, street signs, and other physical objects that provide visual navigation information. The TLD system generates TLD data as an output 504d. The TLD data often takes the form of image data (e.g., data in an image data format such as RAW, JPEG, PNG, etc.). The TLD system differs from a system that includes a camera in that the TLD system uses a camera with a wide field of view (e.g., using a wide-angle lens or a fisheye lens) to obtain information about as many physical objects that provide visual navigation information as possible, allowing the AV 100 to access all relevant navigation information provided by these objects. For example, the viewing angle of the TLD system can be approximately 120 degrees or greater.

[0115] In some embodiments, the outputs 504a-504d are combined using sensor fusion techniques. Thus, the individual outputs 504a-504d are provided to other systems of the AV 100 (e.g., to a Figure 4 The combined output may be provided to other systems in the form of a single combined output or multiple combined outputs of the same type (e.g., using the same combining technique or combining the same outputs, or both) or a single combined output or multiple combined outputs of different types (e.g., using different individual combining techniques or combining different individual outputs, or both). In some embodiments, an early fusion technique is used. An early fusion technique is characterized in that the outputs are combined before one or more data processing steps are applied to the combined output. In some embodiments, a late fusion technique is used. A late fusion technique is characterized in that the outputs are combined after one or more data processing steps are applied to the individual outputs.

[0116] Figure 6 An example of a LiDAR system 602 is shown (e.g., Figure 5Input 502a is shown. The LiDAR system 602 emits light 604a-604c from a light emitter 606 (e.g., a laser emitter). The light emitted by the LiDAR system is typically not in the visible spectrum; for example, infrared light is often used. Some of the emitted light 604b encounters a physical object 608 (e.g., a vehicle) and reflects back to the LiDAR system 602. (The light emitted from the LiDAR system typically does not penetrate a physical object, such as a solid form of physical object.) The LiDAR system 602 also has one or more light detectors 610 for detecting the reflected light. In an embodiment, one or more data processing systems associated with the LiDAR system generate an image 612 representing a field of view 614 of the LiDAR system. The image 612 includes information representing the boundaries 616 of the physical object 608. In this way, the image 612 is used to determine the boundaries 616 of one or more physical objects in the vicinity of the AV.

[0117] Figure 7 1 illustrates a LiDAR system 602 in operation. In the scenario shown in this figure, the AV 100 receives both camera system output 504c in the form of image 702 and LiDAR system output 504a in the form of LiDAR data points 704. In use, the data processing system of the AV 100 compares the image 702 with the data points 704. In particular, the physical objects 706 identified in the image 702 are also identified in the data points 704. In this way, the AV 100 perceives the boundaries of the physical objects based on the contours and density of the data points 704.

[0118] Figure 8 6. As described above, the AV 100 detects the boundaries of physical objects based on the characteristics of the data points detected by the LiDAR system 602. Figure 8 As shown, a flat object such as the ground 802 will reflect light 804a-804d emitted from the LiDAR system 602 in a consistent manner. In other words, because the LiDAR system 602 emits light using consistent intervals, the ground 802 will reflect light back to the LiDAR system 602 at the same consistent intervals. As the AV 100 drives over the ground 802, and nothing is blocking the road, the LiDAR system 602 will continue to detect light reflected by the next valid ground point 806. However, if an object 808 blocks the road, the light 804e-804f emitted by the LiDAR system 602 will reflect from points 810a-810b in a manner that is inconsistent with the expected consistent manner. Based on this information, the AV 100 can determine that the object 808 is present.

[0119] Path Planning

[0120] Figure 9 For example (e.g. Figure 4 900 shows a block diagram of the relationship between the inputs and outputs of the planning module 404 (shown in FIG. 1 ). Generally speaking, the output of the planning module 404 is a route 902 from a starting point 904 (e.g., a source location or initial location) to an end point 906 (e.g., a destination or final location). The route 902 is typically defined by one or more road segments. For example, a road segment refers to a distance to be traveled on at least a portion of a street, road, highway, lane, or other physical area suitable for automobile travel. In some examples, for example, if the AV 100 is an off-road capable vehicle such as a four-wheel drive (4WD) or all-wheel drive (AWD) car, SUV, or pickup truck, the route 902 includes "off-road" segments such as unpaved paths or open fields.

[0121] In addition to route 902, the planning module also outputs lane-level routing data 908. Lane-level routing data 908 is used to navigate the segments of route 902 at a specific time based on the conditions of those segments. For example, if route 902 comprises a multi-lane highway, lane-level routing data 908 includes trajectory planning data 910, which the AV 100 can use to select a lane from the multiple lanes based on, for example, whether an exit is imminent, whether other vehicles are available in one or more of the multiple lanes, or other factors that change over the course of a few minutes or less. Similarly, in some implementations, lane-level routing data 908 includes rate constraints 912 specific to a segment of route 902. For example, if the segment includes pedestrians or unexpected traffic, rate constraints 912 can limit the AV 100 to a slower travel speed than expected, such as a speed based on speed limit data for the segment.

[0122] In an embodiment, inputs to the planning module 404 include (e.g., Figure 4 ) database data 914, current location data 916 (e.g., Figure 4 AV position 418 shown), (e.g., for Figure 4 Destination data 918 and object data 920 (e.g., Figure 4(See the example of a classified object 416 perceived by perception module 402, shown in Figure 4A). In some embodiments, database data 914 includes rules used during planning. Rules are specified using a formal language (e.g., using Boolean logic). In any given situation encountered by AV 100, at least some of these rules will apply to that situation. A rule applies to a given situation if it has conditions that are satisfied based on information available to AV 100 (e.g., information about the surrounding environment). Rules can have priorities. For example, a rule that says, "If the road is a freeway, move to the leftmost lane" may have a lower priority than a rule that says, "If an exit is within one mile, move to the rightmost lane."

[0123] Figure 10 In the example of path planning (eg, by the planning module 404 ( Figure 4 )) uses a directed graph 1000. Generally speaking, if Figure 10 A directed graph 1000, such as the one shown, is used to determine a path between any origin 1002 and destination 1004. In the real world, the distance separating the origin 1002 and destination 1004 may be relatively large (e.g., in two different metropolitan areas) or may be relatively small (e.g., two intersections adjacent to a city block or two lanes of a multi-lane road).

[0124] In one embodiment, directed graph 1000 includes nodes 1006a-1006d representing different locations that AV 100 may occupy between starting point 1002 and end point 1004. In some examples, for example, when starting point 1002 and end point 1004 represent different metropolitan areas, nodes 1006a-1006d represent sections of a road. In some examples, for example, when starting point 1002 and end point 1004 represent different locations on the same road, nodes 1006a-1006d represent different locations on that road. Thus, directed graph 1000 includes information at different levels of granularity. In one embodiment, a directed graph with high granularity is also a subgraph of another directed graph with a larger scale. For example, most of the information for a directed graph where the start point 1002 and the end point 1004 are far apart (e.g., many miles apart) is at a low granularity and is based on stored data, but the directed graph also includes some high granularity information for a portion of the directed graph that represents a physical location in the field of view of the AV 100.

[0125] Nodes 1006a-1006d are distinct from objects 1008a-1008b that cannot overlap with nodes. In an embodiment, at low granularity, objects 1008a-1008b represent areas that cannot be traversed by cars, such as areas without streets or roads. At high granularity, objects 1008a-1008b represent physical objects in the field of view of the AV 100, such as other cars, pedestrians, or other entities with which the AV 100 cannot share physical space. In an embodiment, some or all of objects 1008a-1008b are static objects (e.g., objects that do not change position, such as streetlights or utility poles) or dynamic objects (e.g., objects that can change position, such as pedestrians or other cars).

[0126] Nodes 1006a-1006d are connected by edges 1010a-1010c. If two nodes 1006a-1006b are connected by edge 1010a, the AV 100 can travel between one node 1006a and another node 1006b, for example, without having to travel to an intermediate node before reaching the other node 1006b. (When referring to the AV 100 traveling between nodes, it means that the AV 100 travels between two physical locations represented by the corresponding nodes.) Edges 1010a-1010c are typically bidirectional, in the sense that the AV 100 travels from a first node to a second node, or from a second node to a first node. In an embodiment, edges 1010a-1010c are unidirectional, in the sense that the AV 100 can travel from a first node to a second node, but the AV 100 cannot travel from a second node to the first node. Edges 1010a-1010c are unidirectional where they represent, for example, one-way streets, a single lane of a street, road, or highway, or other features that can only be traversed in one direction due to legal or physical constraints.

[0127] In an embodiment, the planning module 404 uses the directed graph 1000 to identify a path 1012 consisting of nodes and edges between the start point 1002 and the end point 1004 .

[0128] Edges 1010a-1010c have associated costs 1014a-1014b. Costs 1014a-1014b are values that represent resources that will be spent if the AV 100 chooses that edge. A typical resource is time. For example, if one edge 1010a represents a physical distance that is twice that of another edge 1010b, the associated cost 1014a of the first edge 1010a can be twice that of the associated cost 1014b of the second edge 1010b. Other factors that affect time include expected traffic, number of intersections, speed limits, etc. Another typical resource is fuel economy. Two edges 1010a-1010b can represent the same physical distance, but one edge 1010a requires more fuel than the other edge 1010b due to, for example, road conditions, expected weather, etc.

[0129] When the planning module 404 identifies a path 1012 between the start point 1002 and the end point 1004, the planning module 404 typically chooses a path that is optimized for cost, e.g., the path that has the smallest total cost when the individual costs of the edges are added together.

[0130] System and method for verifying calibration of a sensor system

[0131] Figure 13 A system for verifying sensor calibration according to one or more embodiments of the present application is illustrated. Figure 13 The system in FIG. 13 includes a first sensor 1301, a second sensor 1302, an object 1303, and a computer processor 1304. The computer processor 1304 includes a computer readable medium 1305. The computer readable medium 1305 includes computer executable instructions 1306 stored thereon. The object 1303 is configured to have a substantially non-reflective portion 1303a, and in some embodiments a substantially reflective portion 1303b.

[0132] The object 1303 can be configured to have one of a number of types of shapes. For example, in the illustrated embodiment, the object 1303 has a spherical shape. In an embodiment, the object 1303 has a cubical shape. In an embodiment, the object 1303 has a cuboid shape. In an embodiment, the object 1303 has a conical shape. The object 1303 can also have a toroidal shape, a triangular shape, a cylindrical shape, a pyramidal shape, or any of a number of shape primitives. In an embodiment, the object 1303 is a sphere. In an embodiment, the object 1303 is a cone. In an embodiment, the object 1303 is a cube. In an embodiment, the object 1303 is located in a building. In an embodiment, the object 1303 is a fixed roadside feature (e.g., a street sign, a telephone pole, a billboard, a bridge support column, etc.). For example, the object can be a billboard that incorporates a special code, text, image, or graphic that is a priori or a posteriori appropriate for calibration.

[0133] As previously described, object 1303 can be configured to have a substantially non-reflective portion 1303a and a substantially reflective portion 1303b. Substantially non-reflective portion 1303a comprises a material such as glass, paint, fabric, or coating that absorbs most or all incident light (e.g., providing a reflectivity of less than 5%), regardless of the visible spectrum. For example, in one embodiment, substantially non-reflective portion 1303a comprises black paint. In one embodiment, substantially non-reflective portion 1303a comprises black fabric. In one embodiment, substantially non-reflective portion 1303a comprises an anti-reflective coating, which may include a transparent thin film structure with alternating layers that create a contrasting refractive index. Substantially reflective portion 1303b comprises a material such as paint, mirror, fabric, or metal that reflects most or all incident light (e.g., providing a reflectivity of greater than 70%). Materials and objects are not limited to mirrored or Lambertian (or a combination of both). For example, in one embodiment, substantially reflective portion 1303b comprises a glass mirror. In one embodiment, substantially reflective portion 1303b comprises an acrylic mirror. In an embodiment, the substantially reflective portion 1303b comprises reflective tape.In an embodiment, the substantially reflective portion 1303b comprises biaxially oriented polyethylene terephthalate ("Mylar").

[0134] Sensors 1301, 1302 can each be one of several types of sensing devices. For example, in an embodiment, each sensor 1301, 1302 is previously referenced Figure 1 In one embodiment, each sensor 1301, 1302 is as previously described with reference to Figure 5One or more of the inputs 502a-502c discussed. In the illustrated embodiment, the first sensor 1301 is a LiDAR and the second sensor 1302 is a camera. The camera can be a monocular or stereo camera configured to capture light in the visible, infrared, ultraviolet, and / or thermal spectrum. In an embodiment, at least one of the sensors 1301, 1302 is an ultrasonic sensor. In an embodiment, the first sensor 1301 is a RADAR. At least one of the sensors 1301, 1302 can also include a combination of sensing devices. For example, in an embodiment, at least one of the sensors 1301, 1302 includes a camera and a RADAR. In an embodiment, at least one of the sensors 1301, 1302 also includes additional sensors for sensing or measuring properties of the environment of the AV (e.g., AV 100). For example, a monocular or stereo camera 122 capable of sensing light in the visible, infrared, and / or thermal spectrum, a LiDAR 123, a RADAR, an ultrasonic sensor, a time-of-flight (TOF) depth sensor, a velocity sensor, a temperature sensor, a humidity sensor, and a precipitation sensor. Although the illustrated embodiment includes a first sensor 1301 and a second sensor 1302, Figure 13 The system may include one or more additional sensors. For example, in an embodiment, Figure 13 The system includes a third sensor. In an embodiment, Figure 13 The system includes a third, a fourth, and a fifth sensor.

[0135] In an embodiment, the first sensor 1301 is configured to detect the distance to (e.g., one or more) locations associated with the substantially reflective portion 1303b. For example, in an embodiment, the first sensor 1301 transmits rapid pulses of laser light (or radio waves, infrared light, etc.) at several locations associated with the substantially reflective portion 1303b, and for each of the several locations associated with the substantially reflective portion 1303b, the first sensor 1301 measures the amount of time required for each pulse to reflect from each location. Thus, the first sensor 1301 can measure the distance to each location associated with the substantially reflective portion 1303b by utilizing known intrinsic properties of light (e.g., velocity). In an embodiment, the first sensor 1301 does not detect the distance to the location associated with the substantially non-reflective portion 1303a because, for example, the substantially non-reflective portion 1303a does not reflect sufficient light for the first sensor 1301 to detect.

[0136] In an embodiment, the second sensor 1302 is an imaging sensor (e.g., a camera) configured to detect light intensity values ​​at (e.g., one or more) locations associated with the substantially reflective portion 1303b of the object 1303. For example, in an embodiment, the second sensor 1302 includes a CCD and / or CMOS sensor having pixels, wherein each pixel can detect an incident photon from a location associated with the substantially reflective portion 1303b and generate an electron having an energy corresponding to the power (and thus the intensity) of the incident photon. Due to the spatial arrangement of the pixels, the second sensor 1302 can also detect the spatial location of the intensity value. In an embodiment, the second sensor 1302 is configured to also detect light intensity values ​​at (e.g., one or more) locations associated with the substantially non-reflective portion 1303a of the object 1303. Typically, the light intensity values ​​at the locations associated with the substantially non-reflective portion 1303a are smaller than the light intensity values ​​of the substantially reflective portion 1303b.

[0137] The computer-readable medium 1306 (or computer-readable memory) may include any type of data storage technology suitable for the local technical environment, including but not limited to semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory, removable memory, disk memory, flash memory, dynamic random access memory (DRAM), static random access memory (SRAM), and electronically erasable programmable read-only memory (EEPROM), etc. In an embodiment, the computer-readable medium 1306 stores code segments having computer-executable instructions 1306.

[0138] In an embodiment, the computer processor 1304 includes the same Figure 3 The processor 304 discussed above is similar to one or more computer processors (e.g., a microprocessor, a microcontroller, or both). The computer processor 1304 is configured to execute program code, such as computer-executable instructions 1306. The computer processor 1304 is configured to be communicatively coupled to the first sensor 1301 and the second sensor 1302. When the computer processor 1304 executes the computer-executable instructions 1306, the computer processor 1304 performs several operations.

[0139] In an embodiment, when computer processor 1304 is executing computer executable instructions 1306, computer processor 1304 performs operations to receive sensor data associated with a detected distance to a location associated with base reflective portion 1303b of object 1303 from first sensor 1301. Based on the received sensor data, computer processor 1304 also performs operations to generate a first predicted aggregate location associated with base reflective portion 1303b. For example, Figure 14is an illustrative example of generating a first predicted aggregate position 1403a and a second predicted aggregate position 1403b according to one or more embodiments of the present invention. Figure 14 After receiving sensor data associated with a detected distance to a location associated with substantially reflective portion 1303b, computer processor 1304 generates a first predicted aggregation location 1403a for substantially reflective portion 1303b based on the sensor data. In an embodiment, computer processor 1304 determines first predicted aggregation location 1403a using a first algorithm embodied in computer-executable instructions 1306. Typically, because first sensor 1301 does not detect a distance to the location of substantially non-reflective portion 1303a, computer processor 1304 is able to determine the location of substantially reflective portion 1303b, which may facilitate the generation of first predicted aggregation location 1403a.

[0140] Return Reference Figure 13 When executing computer executable instructions 1306, the computer processor performs operations to receive sensor data associated with the intensity value of the position associated with the basic reflective portion 1303b from the second sensor 1302. Based on the sensor data received from the second sensor 1302, the computer processor 1304 also performs operations to generate a second predicted aggregate position associated with the basic reflective portion 1303b. For example, referring again to Figure 14After receiving sensor data associated with intensity values ​​at locations associated with substantially reflective portion 1303b, computer processor 1304 generates a second predicted aggregate position 1403b for substantially reflective portion 1303b based on the sensor data. In an embodiment, computer processor 1304 uses a second algorithm stored in computer-executable instructions 1306 to determine second predicted aggregate position 1403b. In an embodiment, the first algorithm and the second algorithm are different. The first algorithm may generally be based on detected distances, and the second algorithm may be based on detected intensity values. For example, the first algorithm may be based on binomial regression. The second algorithm may include a continuously adaptive mean shift (Camshift) function. Typically, because second sensor 1301 does not detect intensity values ​​at locations associated with substantially non-reflective portion 1303a (or because intensity values ​​at locations associated with substantially non-reflective portion 1303a will be substantially less than intensity values ​​at locations associated with substantially reflective portion 1303b), computer processor 1304 may isolate the location of substantially reflective portion 1303b, which may facilitate the generation of second predicted aggregate position 1403b. Although the first predicted aggregate position 1403a and the second predicted aggregate position 1403b are shown as slightly misaligned relative to each other for illustrative purposes, in reality, the first predicted aggregate position 1403a and the second predicted aggregate position 1403b may be more aligned than shown, less aligned than shown, or completely aligned.

[0141] Return Reference Figure 13 , when executing computer executable instructions 1306, the computer processor 1304 determines an alignment error value based on the first predicted convergence position and the second predicted convergence position. As previously described, the first predicted convergence position and the second predicted convergence position of the basic reflective portion 1303b may be aligned or misaligned relative to each other. Based on the alignment (or misalignment), the computer processor 1304 determines the alignment error. For example, referring to Figure 14 If the first predicted aggregate position 1403a and the second predicted aggregate position 1403b completely overlap (and are therefore substantially aligned), the computer processor 1304 may determine that the alignment error relative to the first predicted aggregate position 1403a and the second predicted aggregate position 1403b is 0%. Generally, the less the first predicted aggregate position 1403a and the second predicted aggregate position 1403b overlap, the greater the alignment error will be.

[0142] As previously described, a first algorithm may be used to determine the first predicted aggregate position 1403a, and a second algorithm may be used to determine the second predicted aggregate position 1403b. In an embodiment, if the alignment error value is greater than a first alignment error threshold, the computer processor 1304 adjusts the first algorithm and / or the second algorithm. The first alignment error threshold may be selected based on safety, desired accuracy, and / or computational efficiency considerations. In an embodiment, the first alignment error threshold is 1%. In an embodiment, the first alignment error threshold is 5%. In an embodiment, if the alignment error is greater than the first alignment error threshold, the computer processor 1304 adjusts the first algorithm and / or the second algorithm so that the first predicted aggregate position and the second predicted aggregate position are substantially aligned (e.g., the alignment error is 0% or significantly close to 0%). In an embodiment, if the alignment error value is greater than the first alignment error threshold, the first algorithm and / or the second algorithm are adjusted until the alignment error value is below the alignment error threshold. In an embodiment, if the alignment error is greater than the second alignment error threshold, the computer processor 1304 initiates a calibration process for the first sensor 1301 and / or the second sensor 1302. In an embodiment, the second alignment error threshold is greater than the first alignment error threshold. The second alignment error threshold may be selected to reflect situations where the first predicted aggregate position 1403a and the second predicted aggregate position 1403b are not sufficiently aligned such that the update algorithm may not be feasible due to safety and / or computational considerations. Figures 17 to 19 An example of the calibration process will be described in detail.

[0143] Figure 15 is a flow chart describing a method for verifying sensor calibration according to one or more embodiments of the present invention. For illustrative purposes, Figure 15 The method in will be described as Figure 13 The system used to verify sensor calibration is carried out in Figure 15 The method may be performed with other systems for verifying sensor calibration. Figure 15 The method in includes: configuring an object to have a reflective portion and a non-reflective portion (block 1510); detecting a distance to one or more locations associated with the reflective portion (block 1511); detecting an intensity value at one or more locations associated with the reflective portion (block 1512); receiving data associated with the distance to the one or more locations associated with the reflective portion (block 1513); generating a first predicted aggregate position (block 1514); receiving data associated with the intensity value at one or more locations associated with the reflective portion (block 1515); generating a second predicted aggregate position (block 1516); and determining an alignment error (block 1517). In an embodiment, Figure 15 The method includes adjusting the algorithm (block 1518). In an embodiment, Figure 15The method includes calibrating the sensor (block 1519).

[0144] At block 1510, the object 1303 is configured to have a substantially non-reflective portion 1303a and a substantially reflective portion 1303b. Figure 13 As shown, substantially non-reflective portion 1303a can be comprised of a material that absorbs most or all incident light (e.g., providing a reflectivity of less than 5%), such as glass, paint, fabric, coating, etc. For example, in an embodiment, substantially non-reflective portion 1303a comprises black paint. In an embodiment, substantially non-reflective portion 1303a comprises black fabric. Substantially non-reflective portion 1303a can also be comprised of an anti-reflective coating, which can include a transparent thin film structure with alternating layers that create a contrasting refractive index. Substantially reflective portion 1303b can be comprised of a material that reflects most or all incident light (e.g., providing a reflectivity of greater than 70%), such as paint, mirror, fabric, metal, etc. For example, in an embodiment, substantially reflective portion 1303b comprises a glass mirror. In an embodiment, substantially reflective portion 1303b comprises an acrylic mirror. In an embodiment, substantially reflective portion 1303b comprises reflective tape. In an embodiment, substantially reflective portion 1303b comprises biaxially oriented polyethylene terephthalate ("Mylar").

[0145] At block 1511, the first sensor 1301 detects the distance to (e.g., one or more) locations associated with the substantially reflective portion 1303b. For example, in an embodiment, the first sensor 1301 is a LiDAR and transmits rapid pulses of laser light at several locations associated with the substantially reflective portion 1303b. For each of the several locations associated with the substantially reflective portion 1303b, the first sensor 1301 measures the amount of time required for each pulse to reflect from each location. Thus, the first sensor 1301 can measure the distance to each location associated with the substantially reflective portion 1303b by utilizing inherent properties of light (e.g., velocity). In an embodiment, the first sensor 1301 does not detect the distance to the location associated with the substantially non-reflective portion 1303a because, for example, the substantially non-reflective portion 1303a does not reflect sufficient light for the first sensor 1301 to detect.

[0146] At block 1512, the second sensor 1302 detects light intensity values ​​at (e.g., one or more) locations associated with the substantially reflective portion 1303b of the object 1303. For example, in an embodiment, the second sensor 1302 includes a CCD and / or CMOS sensor having spatially aligned pixels, wherein each pixel can detect an incident photon from a location associated with the substantially reflective portion 1303b and generate an electron having an energy corresponding to the power (and thus the intensity) of the incident photon. Due to the spatial arrangement of the pixels, the second sensor 1302 can also detect the spatial location of the intensity value. In an embodiment, the second sensor 1302 also detects light intensity values ​​at (e.g., one or more) locations associated with the substantially non-reflective portion 1303b of the object 1303.

[0147] At block 1513 , the computer processor 1304 receives sensor data from the first sensor 1301 associated with detected distances to one or more locations associated with the base reflective portion 1303 b .

[0148] At block 1514, the computer processor 1304 generates a first predicted aggregation position 1403a for the substantially reflective portion 1303b based on the sensor data received from the first sensor 1301. In an embodiment, the computer processor 1304 determines the first predicted aggregation position 1403a using a first algorithm stored in the computer-executable instructions 1306. Typically, because the first sensor 1301 does not detect the distance of the position of the substantially non-reflective portion 1303a, the computer processor 1304 is able to isolate the position of the substantially reflective portion 1303b, which can facilitate the generation of the first predicted aggregation position 1403a.

[0149] At block 1515 , the computer processor 1304 receives sensor data from the second sensor 1302 associated with the detected intensity value of the location associated with the base reflective portion 1303 b .

[0150] At block 1516, the computer processor 1304 generates a second predicted aggregate location 1403b for the substantially reflective portion 1303b based on the sensor data received from the second sensor 1302. In embodiments, the computer processor 1304 determines the second predicted aggregate location 1403b using a second algorithm stored in the computer executable instructions 1306. In embodiments, the first algorithm and the second algorithm are different. For example, the first algorithm can generally be based on detected distances, and the second algorithm can be based on detected intensity values. Generally, because the second sensor 1301 does not detect intensity values for locations associated with the substantially non-reflective portion 1303a (or because the intensity values for locations associated with the substantially non-reflective portion 1303a will be substantially less than the intensity values for locations associated with the substantially reflective portion 1303b), the computer processor 1304 can isolate the locations of the substantially reflective portion 1303b, which can facilitate generation of the second predicted aggregate location 1403b. As shown previously with respect to FIG. 14, the first predicted aggregate location 1403a and the second predicted aggregate location 1403b can be partially aligned, fully aligned, or not aligned at all. Figure 13

[0151] At block 1517, the computer processor 1304 determines an alignment error value based on the first predicted aggregate location 1403a and the second predicted aggregate location 1403b. As shown previously, the first predicted aggregate location 1403a and the second predicted aggregate location 1403b for the substantially reflective portion 1303b can be aligned or not aligned relative to one another. Based on this alignment (or lack thereof), the computer processor 1304 determines an alignment error. For example, if the first predicted aggregate location 1403a and the second predicted aggregate location 1403b overlap completely (and thus are substantially aligned), the computer processor 1304 can determine an alignment error of 0% relative to the first predicted aggregate location 1403a and the second predicted aggregate location 1403b. Generally, the less the first predicted aggregate location 1403a and the second predicted aggregate location 1403b overlap, the greater the alignment error will be.

[0152] ​At block 1518, if the alignment error value is greater than a first alignment error threshold, the computer processor 1304 adjusts the first algorithm and / or the second algorithm used by the computer processor 1304 to determine the first predicted aggregate position 1403a and the second predicted aggregate position 1403b, respectively. The first alignment error threshold may be selected based on safety, desired accuracy, and / or computational efficiency considerations. In an embodiment, the first alignment error threshold is 1%. In an embodiment, the first alignment error threshold is 5%. In an embodiment, if the alignment error is greater than the first alignment error threshold, the first algorithm and / or the second algorithm are adjusted so that the first predicted aggregate position and the second predicted aggregate position are substantially aligned (e.g., the alignment error is 0% or significantly close to 0%). In an embodiment, if the alignment error value is greater than the first alignment error threshold, the first algorithm and / or the second algorithm are adjusted until the alignment error value is below the alignment error threshold.

[0153] At block 1519, if the alignment error is greater than a second alignment error threshold, the computer processor 1304 initiates a calibration process for the first sensor 1301 and / or the second sensor 1302. In an embodiment, the second alignment error threshold is greater than the first alignment error threshold. The second alignment error threshold may be selected to reflect a situation where the first predicted aggregate position 1403a and the second predicted aggregate position 1403b are not sufficiently aligned such that an update algorithm may not be feasible due to safety and / or computational considerations. Figures 17 to 19 An example of the calibration process will be described in detail.

[0154] System and method for calibrating a sensor system

[0155] Figure 16 A system for calibrating sensors according to one or more embodiments of the present invention is illustrated. The system includes an AV 1610, a first sensor 1611a, a second sensor 1611b, a third sensor 1611c, an object 1620, and a computer processor 1630. The computer processor 1630 includes a computer-readable medium 1631. The computer-readable medium 1631 includes computer-executable instructions 1632 stored thereon.

[0156] Object 1620 can be configured to have one of several types of shapes. For example, in the illustrated embodiment, object 1620 has a spherical shape. In an embodiment, object 1620 has a cubic shape. In an embodiment, object 1620 has a cuboidal shape. In an embodiment, object 1620 has a conical shape. Object 1620 may also have a toroidal shape, a triangular shape, a cylindrical shape, a pyramidal shape, and the like. In an embodiment, object 1620 is a sphere. In an embodiment, object 1620 is a cone. In an embodiment, object 1620 is a cube. In an embodiment, object 1620 includes three faces defining mutually perpendicular planes. Object 1620 may also be a fixed road or building structure in the environment of AV 1610. For example, in an embodiment, object 1620 is a street sign. In an embodiment, object 1620 is a billboard. In an embodiment, object 1620 is a building support beam (e.g., a parking lot support beam). Object 1620 may also naturally exist in the environment of AV 1610. For example, in an embodiment, object 1620 is a tree. In an embodiment, object 1620 is a boulder. Although only one object 1620 is shown for illustrative purposes, Figure 16 The system in may include additional objects, each additional object being one of several types (ie, cube, support beam, tree, etc.).

[0157] Object 1620 has at least one calibration feature. For example, in an embodiment, object 1620 has at least one fixed printed pattern detectable by sensors 1611a, 1611b, and 1611c. In an embodiment, object 1620 has a fixed location. In an embodiment, object 1620 is a fixed distance from another object. In an embodiment, object 1620 has multiple sides. In an embodiment, object 1620 has an internal metal core whose size is optimized for RADAR detection (e.g., optimized to reduce uncertainty), and a reinforced portion that is detectable by LiDAR but substantially undetectable by RADAR (e.g., invisible RADAR foam, cardboard, etc.). Thus, object 1620 can be detected by both RADAR and LiDAR while reducing the uncertainty of the measurement from RADAR detection.

[0158] In this embodiment, the first sensor 1611a, the second sensor 1611b, and the third sensor 1611c are mounted on the AV 1610. The AV 1610 may include more or fewer sensors than the first sensor 1611a, the second sensor 1611b, and the third sensor 1611c. For example, in an embodiment, the AV 1610 includes only the first sensor 1611a and the second sensor 1611b. In an embodiment, the AV 1610 includes a fourth sensor (or a fifth sensor and a sixth sensor, etc.).

[0159] Sensors 1611a, 1611b, 1611c can each be one of several types of sensing devices. For example, in an embodiment, sensors 1611a, 1611b, 1611c can each be a previously referenced Figure 1 One of the sensors 121 discussed. In an embodiment, each of the sensors 1611a, 1611b, 1611c is as previously described with reference to Figure 5 One or more of the inputs 502a-502c discussed. In the illustrated embodiment, the first sensor 1611a is a LiDAR, the second sensor 1611b is a camera, and the third sensor 1611c is a RADAR. The camera can be a monocular or stereo camera configured to capture light in the visible, infrared, and / or thermal spectrum. In an embodiment, at least one of the sensors 1611a, 1611b, 1611c is an ultrasonic sensor. At least one of the sensors 1611a, 1611b, 1611c can also include a combination of sensing devices. For example, in an embodiment, at least one of the sensors 1611a, 1611b, 1611c includes a camera and RADAR. In an embodiment, at least one of the sensors 1611a, 1611b, 1611c also includes additional sensors for sensing or measuring properties of the environment of the AV 1610. For example, monocular or stereo cameras 122, LiDAR 123, RADAR, ultrasonic sensors, time-of-flight (TOF) depth sensors, velocity sensors, temperature sensors, humidity sensors, and precipitation sensors in the visible, infrared, or thermal (or both) spectrum.

[0160] Each of the sensors 1611a, 1611b, and 1611c has a plurality of internal parameters. These internal parameters may include, for example, operating frequency (e.g., signal frequency), beam width, signal-to-noise ratio, internal noise, beam intensity level, operating temperature, focal length, and field of view. Each of the sensors 1611a, 1611b, and 1611c has a plurality of external parameters. These external parameters may include, for example, the position of the sensors 1611a, 1611b, and 1611c when mounted on the AV 1610, the distance between the sensors 1611a, 1611b, and 1611c, the noise level of the AV 1610's environment (e.g., fog, smoke, rain, etc.), the ambient brightness of the AV 1610's environment, and the angle / tilt of the sensors 1611a, 1611b, and 1611c. In an embodiment, one or more of the sensors 1611a, 1611b, and 1611c include at least one monitoring device configured to detect the internal parameters of the corresponding sensor. For example, in an embodiment, at least one of the sensors 1611a, 1611b, 1611c comprises a temperature sensor configured to measure an operating temperature. In an embodiment, at least one of the sensors 1611a, 1611b, 1611c comprises an electromagnetic wave sensor capable of measuring electromagnetic power.

[0161] Each of the sensors 1611a, 1611b, and 1611c is configured to detect at least one calibration feature of the object 1620 as the object 1620 and the AV 1610 are moved relative to each other. For example, in one embodiment, the first sensor 1611a is a LiDAR, the second sensor 1611b is a stereo camera, and the third sensor 1611c is a RADAR. The object 1620 is a cuboidal structure having printed patterns (i.e., calibration features) on each of its faces. Each of the sensors 1611a, 1611b, and 1611c is configured to detect the position of each printed pattern of the object 1620 as the object 1620 is rotated around the AV 1610 (e.g., by a person or a machine). As the object 1620 and the AV 1610 are rotated relative to each other, the sensors 1611a, 1611b, and 1611c each generate a plurality of point clouds associated with the printed patterns. In an embodiment, the object 1620 is a stationary road sign, and the sensors 1611a, 1611b, 1611c are each configured to detect the center and / or edges of the road sign (i.e., calibration features) while generating a number of point clouds associated with the center and / or edges of the stationary road sign as the AV 1610 and the road sign move relative to each other.

[0162] Computer-readable media 1631 (or computer-readable memory) may include any type of data storage technology suitable for the local technical environment, including but not limited to semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems, fixed memory, removable memory, disk memory, flash memory, dynamic random access memory (DRAM), static random access memory (SRAM), and electronically erasable programmable read-only memory (EEPROM). In an embodiment, computer-readable media 1631 stores code segments having computer-executable instructions 1632.

[0163] In an embodiment, the computer processor 1630 includes the same Figure 3 The computer processor 1630 is configured to execute program code, such as computer-executable instructions 1632, similar to the processor 304 discussed above. The computer processor 1630 is configured to be communicatively coupled to the first sensor 1611a, the second sensor 1611b, and the third sensor 1611c. In an embodiment, the computer processor 1630 is communicatively coupled to the remote database 134. When the computer processor 1630 executes the computer-executable instructions 1632, the computer processor 1630 performs several operations.

[0164] In an embodiment, when computer processor 1630 is executing computer-executable instructions 1632, computer processor 1630 performs operations to receive known calibration feature data associated with at least one calibration feature of object 1620. The known calibration feature data can be received from a number of sources. For example, in an embodiment, the known calibration feature data is received from a user input. In an embodiment, the known calibration feature data is received from a remote database 134 and corresponds to sensor data from a calibrated sensor remote from AV 1610. For example, assume that object 1620 is a street sign along a road. Computer processor 1630 can receive point cloud data associated with the location of the center of the street sign (or a corner of the street sign) generated by one or more remote sensors (e.g., sensors mounted on a second AV) from remote database 134.

[0165] In an embodiment, known calibration feature data is received from one of sensors 1611a, 1611b, and 1611c of AV 1610 that is known to be the most accurate. For example, assume that object 1620 is a cubic structure with a printed pattern, and first sensor 1611a is known to be the most accurate. In an embodiment, one of sensors 1611a, 1611b, and 1611c has been pre-calibrated, and known calibration feature data is received from this pre-calibrated sensor. In an embodiment, known calibration feature data is received from one of sensors 1611a, 1611b, and 1611c based on the sensor type. For example, if first sensor 1611a is a LiDAR, first sensor 1611a may be selected to represent the known calibration feature data because it may be more accurate at detecting objects located at longer distances. Computer processor 1630 may receive point cloud data associated with the location of the printed pattern from first sensor 1611a. In an embodiment, known calibration data is received from remote database 134 and corresponds to sensor data from at least one of sensors 1611a, 1611b, 1611c according to a previous calibration process. For example, assume that one of sensors 1611a, 1611b, 1611c was calibrated at an earlier time using a cubic structure having a printed pattern. Point cloud data generated during the previous calibration process is stored in remote database 134 (or computer-readable medium 1631), and computer processor 1630 receives the previously generated point cloud data from remote database 134 (or computer-readable medium 1631).

[0166] When computer processor 1630 is executing computer-executable instructions 1632, computer processor 1630 further performs operations to receive, from each of sensors 1611a, 1611b, and 1611c, detected feature data associated with at least one calibration feature of object 1620. For example, in an embodiment, object 1620 is a cubic structure having a printed pattern, and computer processor 1630 receives, from each of sensors 1611a, 1611b, and 1611c, point cloud data associated with the position of the printed pattern as object 1620 and AV 1610 move relative to each other. In an embodiment, object 1620 is a street sign, and computer processor 1630 receives, from each of sensors 1611a, 1611b, and 1611c, point cloud data associated with the position of the center and / or corners of the street sign as the street sign and AV 1610 move relative to each other.

[0167] While the computer processor 1630 is executing the computer-executable instructions 1632, the computer processor 1630 further performs operations to compare the received known feature data to detected feature data received from each of the sensors 1611a, 1611b, 1611c. For example, in an embodiment, the received known feature data is point cloud data associated with a location of a center of a street sign detected by a remote sensor as the second AV moves relative to the street sign (the remote sensor mounted on the AV). The received detected feature data is point cloud data associated with a location of a center of the same street sign detected by each of the sensors 1611a, 1611b, 1611c as the AV 1610 moves relative to the street sign. The computer processor 1630 compares the point cloud data from the remote sensor to the point cloud data from each of the sensors 1611a, 1611b, 1611c. In another embodiment, the received known feature data is point cloud data associated with a location of a printed pattern on a cubical structure generated by a first sensor 1611a that is predetermined to be the most accurate of the three sensors as the cubical structure and the AV 1610 move relative to each other. The received detected feature data is point cloud data associated with a location of the printed pattern on the same cubical structure generated by a second sensor 1611b and a third sensor 1611c as the cubical structure and the AV 1610 move relative to each other. The computer processor 1630 compares the point cloud data generated by the first sensor 1611a to the point cloud data generated by the other sensors 1611b, 1611c.

[0168] When the computer processor 1630 is executing the computer-executable instructions 1632, the computer processor 1630 further performs operations to generate a calibration error value for each of the sensors 1611a, 1611b, and 1611c based on a comparison of the received known feature data with the detected feature data received from the sensors 1611a, 1611b, and 1611c. For example, in an embodiment, the computer processor 1630 uses the known feature data (e.g., point cloud data generated by a remote sensor that is the most accurate sensor among the three sensors 1611a, 1611b, and 1611c) as a reference and compares the detected feature data (e.g., point cloud data) generated by the sensors 1611a, 1611b, and 1611c. In an embodiment, a computer processor compares each point in the point cloud generated by sensors 1611a, 1611b, 1611c to a corresponding point in the reference point cloud and calculates an error for each sensor 1611a, 1611b, 1611c based on the distance between the two. Generally, a smaller distance between each point in the point cloud generated by each of sensors 1611a, 1611b, 1611c and the corresponding point in the reference point cloud will result in a smaller calculated error. Errors may also exist if the reference point cloud contains more points than the point cloud associated with the detected feature data. For example, if object 1620 is at a distance that is fully detectable by LiDAR but may not be fully detectable by the camera at its current focus setting, the LiDAR may generate more points in the point cloud than the camera.

[0169] When computer processor 1630 is executing computer-executable instructions 1632, computer processor 1630 further performs operations to determine that sensors 1611a, 1611b, 1611c are incorrectly calibrated if the corresponding calibration error values ​​of sensors 1611a, 1611b, 1611c exceed a first calibration error threshold. The first calibration error threshold may be selected based on, for example, accuracy, safety, and efficiency considerations. For example, in an embodiment, computer processor 1630 determines that one or more sensors 1611a, 1611b, 1611c are incorrectly calibrated if the corresponding calibration error values ​​of one or more sensors exceed a 5% error. In an embodiment, computer processor 1630 determines that one or more sensors are incorrectly calibrated if points in the corresponding point cloud of one or more sensors 1611a, 1611b, 1611c are separated from corresponding points in the reference point cloud by a distance exceeding 0.5 meters.

[0170] In an embodiment, when computer processor 1630 is executing computer-executable instructions 1632, computer processor 1630 further performs operations to calculate a correction parameter for at least one internal parameter associated with one or more sensors 1611a, 1611b, 1611c determined to be incorrectly calibrated. For example, in an embodiment, computer processor 1630 determines, based on the calibration error value, that the incorrectly calibrated sensor should increase its beam power to increase detection accuracy. In an embodiment, first sensor 1611a is a reference sensor and is a LiDAR, and second sensor 1611b is a camera. If computer processor 1630 determines that the calibration error is caused by object 1620 being too far away for second sensor 1611b to detect in its current state, resulting in fewer points in the point cloud relative to the point cloud of first sensor 1611a, computer processor 1630 determines that the focus of second sensor 1611b needs to be adjusted so that it aligns with the detection range of first sensor 1611a.

[0171] As previously described, sensors 1611a, 1611b, and 1611c may include a monitoring device configured to detect one or more internal parameters of sensors 1611a, 1611b, and 1611c. In an embodiment, computer processor 1630 receives internal data associated with one or more internal parameters of sensors 1611a, 1611b, and 1611c and calculates correction parameters based on the received internal data. For example, in an embodiment, second sensor 1611b is a LiDAR and includes a temperature sensor and an electromagnetic wave sensor. Computer processor 1630 receives the current operating temperature and current beam power setting from the monitoring device of second sensor 1611b. If computer processor 1630 determines that second sensor 1611b is incorrectly calibrated and needs to have its beam power increased (which may increase the operating temperature of second sensor 1611b), computer processor 1630 may calculate the increased beam power based on the received internal data so that the increased beam power does not overheat second sensor 1611b.

[0172] In an embodiment, once the computer processor 1630 calculates the correction parameters, the computer processor 1630 modifies one or more sensors 1611a, 1611b, 1611c that are determined to be incorrectly calibrated according to the calculated correction parameters. For example, if the computer processor 1630 calculates an increased beam power for the second sensor 1611b, the computer processor 1630 may modify the beam power of the second sensor 1611b by, for example, using the beam power adjustment system of the second sensor 1611b. If the computer processor 1630 calculates an increased focal length for the third sensor 1611c, the computer processor 1630 may modify the focal length of the third sensor 1611c by, for example, adjusting at least one lens of the third sensor 1611c.

[0173] In an embodiment, when computer processor 1630 is executing computer-executable instructions 1632, computer processor 1630 further performs operations to determine that one or more of sensors 1611a, 1611b, and 1611c that are determined to be incorrectly calibrated are in a faulty state if the corresponding calibration error value of the one or more sensors exceeds a second calibration error threshold. For example, assume that points in the point cloud associated with detected feature data received from second sensor 1611b are separated from corresponding points in the reference point cloud by a distance of 5 meters. In an embodiment, the second calibration error threshold is 4 meters, and computer processor 1630 determines that second sensor 1611b is in a faulty state. As another example, third sensor 1611c may not generate the same number of points in its point cloud relative to object 1620 as the reference point cloud, resulting in a 20% error. If the second calibration error threshold is 15%, computer processor 1630 may determine that third sensor 1611c is in a faulty state. The fault condition may indicate that the sensor has physically lost alignment, that the sensor is operating in environmental conditions that are unfavorable for detection (e.g., the camera is operating in highly obscured conditions), that the sensor's aperture has become too dirty for detection, etc. In an embodiment, the second calibration threshold is greater than the first calibration error threshold. The second calibration error threshold may be selected based on, for example, accuracy, safety, and efficiency considerations. In an embodiment, the second calibration error threshold is selected to reflect errors that cannot be adequately fixed by fine-tuning internal parameters relative to safety considerations.

[0174] In embodiments, while the computer processor 1630 is executing the computer executable instructions 1632, the computer processor 1630 further performs operations to calculate a correction parameter for at least one extrinsic parameter associated with a sensor 1611a, 1611b, 1611c determined to be in a fault state. For example, assume that the first sensor 1611a is determined to be in a fault state because its point cloud is misaligned with the reference point cloud such that a calibration error value of 20% is calculated. In embodiments, the computer processor 1630 determines, based on the calibration error value, that the tilt angle of the first sensor 1611a should be adjusted to align its point cloud with the reference point cloud.

[0175] In embodiments, while the computer processor 1630 is executing the computer executable instructions 1632, the computer processor 1630 further performs operations to cause one or more vehicles to operate when a calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a calibration error threshold. For example, in embodiments, if a calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a second calibration error threshold, the computer processor 1630 causes the AV 1610 to stop operating (e.g., by shutting down the engine and / or causing the AV 1610 to come to a stop). In embodiments, if a calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a first calibration error threshold, the computer processor 1630 performs operations to notify a remote technician. In embodiments, if a calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a third calibration error threshold, the computer processor 1630 performs operations to cause the AV 1610 to navigate to a service location (e.g., by updating the route previously discussed with reference to the planning module 404). Figure 4 In embodiments, if a calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a first calibration error threshold, the computer processor 1630 performs operations to disable (e.g., power off) the sensor whose calibration error value exceeds the first calibration error threshold. The calibration error thresholds associated with various vehicle operations can be selected based on safety, accuracy, and efficiency considerations.

[0176] Figure 17 is a flowchart that describes a method for calibrating sensors in accordance with one or more embodiments of the present application. For illustrative purposes, the method is described as being performed by the planning module 404 previously discussed with reference to Figure 16 Figure 16 ​The method is performed by a system for calibrating sensors in a system for calibrating sensors. However, the method can be performed by other systems for calibrating sensors. The method includes: moving a vehicle and an object relative to each other (block 1701); detecting a calibration feature (block 1702); receiving known calibration feature data (block 1703); receiving detected feature data (block 1704); and generating a calibration error (block 1705).

[0177] At block 1701, object 1620 and AV 1610 move relative to each other. In an embodiment, object 1620 rotates around AV 1610 while AV 1610 remains stationary. In an embodiment, object 1620 is stationary, and AV 1610 rotates around object 1620. In an embodiment, object 1620 is a stationary road fixture, and AV 1610 drives past object 1620. In an embodiment, object 1620 has a spherical shape. In an embodiment, object 1620 has a cubic shape. In an embodiment, object 1620 has a cuboidal shape. In an embodiment, object 1620 has a conical shape. Object 1620 may also have a toroidal, triangular, cylindrical, pyramidal, or other shapes. In an embodiment, object 1620 is a sphere. In an embodiment, object 1620 is a cone. In an embodiment, object 1620 is a cube. In an embodiment, object 1620 includes three faces defining mutually perpendicular planes. Object 1620 may also be a road or building structure in the environment of AV 1610. For example, in an embodiment, object 1620 is a street sign. In an embodiment, object 1620 is a billboard. In an embodiment, object 1620 is a building support beam (e.g., a parking lot support beam). Object 1620 may also exist naturally in the environment of AV 1610. For example, in an embodiment, object 1620 is a tree. In an embodiment, object 1620 is a boulder. Although only one object 1620 is shown for illustrative purposes, Figure 16 The system in may include additional objects, each additional object being one of several types (ie, cube, support beam, tree, etc.).

[0178] At block 1702, as the object 1620 and the AV 1610 move relative to each other, the sensors 1611a, 1611b, 1611c detect at least one calibration feature of the object 1620. Figure 16As shown, object 1620 may have at least one calibration feature. For example, in an embodiment, object 1620 has at least one fixed printed pattern detectable by sensors 1611a, 1611b, and 1611c. In an embodiment, object 1620 has a fixed position. In an embodiment, object 1620 has a fixed distance from another object. In an embodiment, object 1620 has multiple edges. In an embodiment, object 1620 has an internal metal core whose size is optimized for RADAR detection (e.g., optimized to reduce uncertainty), and a reinforced portion that is detectable by LiDAR but substantially undetectable by RADAR (e.g., invisible RADAR foam, cardboard, etc.). Therefore, object 1620 can be detected by both RADAR and LiDAR while reducing the uncertainty of the measurement from RADAR detection.

[0179] In an embodiment, the first sensor 1611a is a LiDAR, the second sensor 1611b is a stereo camera, the third sensor 1611c is a RADAR, and the object 1610 is a cuboidal structure having printed patterns (i.e., calibration features) on each of its faces. The sensors 1611a, 1611b, and 1611c are each configured to detect the position of each printed pattern on the object 1620 when the object 1620 is rotated around the AV 1610 (e.g., by a person or a machine). As the object 1620 and the AV 1610 are rotated relative to each other, the sensors 1611a, 1611b, and 1611c each generate a plurality of point clouds associated with the printed patterns. In an embodiment, the object 1610 is a stationary road sign, and the sensors 1611a, 1611b, 1611c are each configured to detect the center and / or edges of the road sign (i.e., calibration features) while generating a number of point clouds associated with the center and / or edges of the stationary road sign as the AV 1610 and the road sign move relative to each other.

[0180] At block 1703, computer processor 1630 receives known calibration feature data associated with calibration features of object 1620. The known calibration feature data can be received from a number of sources. For example, in an embodiment, the known calibration feature data is received from user input. In an embodiment, the known calibration feature data is received from remote database 134 and corresponds to sensor data from a calibrated sensor remote from AV 1610. For example, assume that object 1620 is a street sign along a road. Computer processor 1630 can receive point cloud data generated by one or more remote sensors (e.g., sensors mounted on a second AV) from remote database 134 associated with the location of the center of the street sign (or a corner of the street sign).

[0181] In an embodiment, known calibration feature data is received from the sensor known to be most accurate among sensors 1611a, 1611b, and 1611c of AV 1610. For example, assume that object 1620 is a cubic structure with a printed pattern, and first sensor 1611a is known to be the most accurate. In an embodiment, one of sensors 1611a, 1611b, and 1611c has been pre-calibrated, and known calibration feature data is received from this pre-calibrated sensor. In an embodiment, known calibration feature data is received from one of sensors 1611a, 1611b, and 1611c based on the sensor type. For example, if first sensor 1611a is a LiDAR, first sensor 1611a may be selected to represent the known calibration feature data because it may be more accurate at detecting objects located at longer distances. Computer processor 1630 may receive point cloud data associated with the location of the printed pattern from first sensor 1611a. In an embodiment, known calibration data is received from remote database 134 and corresponds to sensor data from at least one of sensors 1611a, 1611b, and 1611c according to a previous calibration process. For example, assume that one of sensors 1611a, 1611b, and 1611c was calibrated at an earlier time using a cubic structure having a printed pattern. Point cloud data generated during the previous calibration process is stored in remote database 134 (or computer-readable medium 1631), and computer processor 1630 receives the previously generated point cloud data from remote database 134 (or computer-readable medium 1631).

[0182] At block 1704, the computer processor 1630 receives detected feature data associated with calibration features of the object 1620 from the sensors 1611a, 1611b, and 1611c. For example, in an embodiment, the object 1620 is a cubic structure having a printed pattern, and the computer processor 1630 receives point cloud data associated with the position of the printed pattern as the object 1620 and the AV 1610 move relative to each other from each of the sensors 1611a, 1611b, and 1611c. In an embodiment, the object 1620 is a street sign, and the computer processor 1630 receives point cloud data associated with the position of the center and / or corners of the street sign as the street sign and the AV 1610 move relative to each other from each of the sensors 1611a, 1611b, and 1611c.

[0183] At block 1705, the computer processor generates calibration errors by comparing known calibration feature data to detected feature data received from individual ones of the sensors 1611a, 1611b, 1611c. For example, in an embodiment, the received known feature data is point cloud data associated with the locations of the center and / or corners of a street sign detected by a remote sensor as the second AV moves relative to the street sign (the remote sensor is mounted on the AV). The received detected feature data is point cloud data associated with the locations of the center and / or corners of the same street sign detected by each of the sensors 1611a, 1611b, 1611c as the AV 1610 moves relative to the street sign. The computer processor 1630 compares the point cloud data from the remote sensor to the point cloud data from individual ones of the sensors 1611a, 1611b, 1611c. In another embodiment, the received known feature data is point cloud data associated with the locations of a printed pattern on a cuboid structure generated by a first sensor 1611a that is predetermined to be the most accurate of the three sensors as the cuboid structure and the AV 1610 move relative to each other. The received detected feature data is point cloud data associated with the locations of the printed pattern on the same cuboid structure generated by a second sensor 1611b and a third sensor 1611c as the cuboid structure and the AV 1610 move relative to each other. The computer processor 1630 compares the point cloud data generated by the first sensor 1611a to the point cloud data generated by the other sensors 1611b, 1611c.

[0184] After comparing the known calibration feature data to the received detected feature data, the computer processor 1630 generates calibration error values for individual ones of the sensors 1611a, 1611b, 1611c based on the comparison of the received known feature data to the detected feature data received from the sensors 1611a, 1611b, 1611c. For example, in an embodiment, the computer processor 1630 compares individual points in the point cloud from the sensors 1611a, 1611b, 1611c to corresponding points in a reference point cloud and calculates an error for each of the sensors 1611a, 1611b, 1611c based on the distance between the two. Generally, a smaller distance between individual points in the point cloud generated by each of the sensors 1611a, 1611b, 1611c and corresponding points in the reference point cloud will result in a smaller calculated error. There can also be an error if the reference point cloud encompasses more points than the point cloud associated with the detected feature data. For example, in a case where the object 1620 is at a distance that can be fully detected by the LiDAR but can not be fully detected by the camera at its current focal length setting, the LiDAR can generate more points in the point cloud than the camera.

[0185] At block 1706, if the corresponding calibration error value of one or more of sensors 1611a, 1611b, 1611c exceeds a first calibration error threshold, computer processor 1630 determines that the one or more sensors are incorrectly calibrated. The first calibration error threshold may be selected based on, for example, accuracy, safety, and efficiency considerations. For example, in an embodiment, if the corresponding calibration error value of one or more of sensors 1611a, 1611b, 1611c exceeds a 5% error, computer processor 1630 determines that the one or more sensors are incorrectly calibrated. In an embodiment, if points in the corresponding point cloud of one or more of sensors 1611a, 1611b, 1611c are separated from corresponding points in the reference point cloud by a distance exceeding 0.5 m, computer processor 1630 determines that the one or more sensors are incorrectly calibrated.

[0186] Figure 18 is a flow chart describing a method for modifying internal parameters based on calibration errors according to one or more embodiments of the present invention. For illustrative purposes, the method is described as being performed by the previously referenced Figure 16 Discussed Figure 16 However, the method may be performed by other systems for calibrating sensors. The method comprises: performing the Figure 17 Discussed Figure 17 The method for calibrating a sensor in (block 1801) further comprises calculating correction parameters (block 1802) and modifying internal parameters (block 1803).

[0187] At block 1801, Figure 17 A method for calibrating sensors in is provided to determine whether one or more of the sensors 1611a, 1611b, 1611c are incorrectly calibrated.

[0188] At block 1802, computer processor 1630 calculates a correction parameter for at least one internal parameter of each of sensors 1611a, 1611b, and 1611c that is determined to be incorrectly calibrated. For example, in one embodiment, computer processor 1630 determines, based on the calibration error value, that the incorrectly calibrated sensor should increase its beam power to improve detection accuracy. In one embodiment, first sensor 1611a is a reference sensor and is a LiDAR, and second sensor 1611b is a camera. If computer processor 1630 determines that the calibration error is caused by object 1620 being too far away for second sensor 1611b to detect in its current state, resulting in fewer points in the point cloud relative to the point cloud of first sensor 1611a, computer processor 1630 determines that the focus of second sensor 1611b needs to be adjusted so that it aligns with the detection range of first sensor 1611a.

[0189] As previously described, sensors 1611a, 1611b, and 1611c may include a monitoring device configured to detect one or more internal parameters of sensors 1611a, 1611b, and 1611c. In an embodiment, computer processor 1630 receives internal data associated with one or more internal parameters of sensors 1611a, 1611b, and 1611c and calculates correction parameters based on the received internal data. For example, in an embodiment, second sensor 1611b is a LiDAR and includes a temperature sensor and an electromagnetic wave sensor. Computer processor 1630 receives the current operating temperature and current beam power setting from the monitoring device of second sensor 1611b. If computer processor 1630 determines that second sensor 1611b is incorrectly calibrated and needs to have its beam power increased (which may increase the operating temperature of second sensor 1611b), computer processor 1630 may calculate the increased beam power based on the received internal data so that the increased beam power does not overheat second sensor 1611b.

[0190] In block 1803, the computer processor 1630 modifies one or more sensors 1611a, 1611b, 1611c that are determined to be incorrectly calibrated based on the calculated correction parameters. For example, if the computer processor 1630 calculates an increased beam power for the second sensor 1611b, the computer processor 1630 may modify the beam power of the second sensor 1611b by, for example, using the beam power adjustment system of the second sensor 1611b. If the computer processor 1630 calculates an increased focal length for the third sensor 1611c, the computer processor 1630 may modify the focal length of the second sensor 1611c by, for example, adjusting at least one lens of the third sensor 1611c.

[0191] Figure 19 is a flow chart describing a method for determining sensor failure based on calibration error according to one or more embodiments of the present invention. For the purpose of illustration, the method is described as being Figure 16 Discussed Figure 16 However, the method may be performed by other systems for calibrating sensors. The method comprises: performing the above-mentioned Figure 17 Discussed Figure 17 The method for calibrating a sensor in a vehicle (block 1901) further includes determining a sensor fault based on a calibration error (block 1902), operating a vehicle based on the calibration error (block 1903), and calculating a correction parameter for an external parameter (block 1904).

[0192] At block 1901 , a method for calibrating sensors is performed to determine whether one or more of the sensors 1611a , 1611b , 1611c are incorrectly calibrated.

[0193] At block 1902, computer processor 1630 further performs operations to determine that one or more of sensors 1611a, 1611b, and 1611c that are determined to be incorrectly calibrated are in a faulty state if the corresponding calibration error value of the one or more sensors exceeds a second calibration error threshold. For example, assume that points in the point cloud associated with detected feature data received from second sensor 1611b are separated from corresponding points in the reference point cloud by a distance of 5 meters. In an embodiment, the second calibration error threshold is 4 meters, and computer processor 1630 determines that second sensor 1611b is in a faulty state. As another example, third sensor 1611b may not generate the same number of points in its point cloud relative to object 1620 as the reference point cloud, resulting in a 20% error. If the second calibration error threshold is 15%, computer processor 1630 may determine that third sensor 1611b is in a faulty state. The fault condition may indicate that the sensor has physically lost alignment, that the sensor is operating in environmental conditions that are unfavorable for detection (e.g., the camera is operating in highly obscured conditions), that the sensor's aperture has become too dirty for detection, etc. In an embodiment, the second calibration threshold is greater than the first calibration error threshold. The second calibration error threshold may be selected based on, for example, accuracy, safety, and efficiency considerations. In an embodiment, the second calibration error threshold is selected to reflect errors that cannot be adequately fixed by fine-tuning internal parameters relative to safety considerations.

[0194] At block 1903, the computer processor 1630 further performs operations to cause one or more vehicles to operate if the calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a calibration error threshold. For example, in an embodiment, if the calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a second calibration error threshold, the computer processor 1630 causes the AV 1610 to stop operating (e.g., by shutting down the engine and / or stopping the AV 1610). In an embodiment, if the calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a first calibration error threshold, the computer processor 1630 performs operations to notify a remote technician. In an embodiment, if the calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a third calibration error threshold, the computer processor 1630 performs operations to (e.g., by updating a previously referenced Figure 4 The planning module 404 discussed above navigates the vehicle to the service location. In an embodiment, if the calibration error value of at least one of the sensors 1611a, 1611b, 1611c exceeds a first calibration error threshold, the computer processor 1630 performs an operation to disable (e.g., power off) the sensor whose calibration error value exceeds the first calibration error threshold. The calibration error threshold associated with each vehicle operation can be selected based on safety, accuracy, and efficiency considerations.

[0195] At block 1904, computer processor 1630 performs operations to calculate a correction parameter for at least one external parameter associated with sensor 1611a, 1611b, or 1611c that is determined to be in a faulty state. For example, assume that first sensor 1611a is determined to be in a faulty state because its point cloud is misaligned with the reference point cloud, resulting in a calculated calibration error value of 20%. In one embodiment, computer processor 1630 determines that the tilt angle of first sensor 1611a should be adjusted based on the calibration error value to align its point cloud with the reference point cloud.

[0196] Additional embodiments

[0197] In an embodiment, a system includes: a vehicle; and at least one object including at least one calibration feature. The vehicle includes a sensor configured to detect the at least one calibration feature of the at least one object as the at least one object and the vehicle move relative to each other. Each sensor has a plurality of parameters. The parameters include internal parameters and external parameters. A computer-readable medium stores computer-executable instructions. At least one processor is configured to communicatively couple to the sensor and execute the instructions stored on the computer-readable medium. The at least one processor executes the instructions and performs operations to receive known calibration feature data associated with the at least one calibration feature. Detected feature data associated with the at least one calibration feature is received from each sensor. The received known feature data is compared with the received detected feature data. For each sensor, a calibration error value is generated based on the comparison of the received known feature data with the received detected feature data. If the calibration error value corresponding to at least one sensor is greater than a first calibration error threshold, it is determined that the at least one sensor is incorrectly calibrated.

[0198] In an embodiment, the sensor comprises a first sensor of a first type and a second sensor of a second type.

[0199] In an embodiment, the sensor comprises at least one RADAR sensor, and the at least one object comprises an internal metal core detectable by the at least one RADAR sensor.

[0200] In an embodiment, the sensor includes at least one light detection and ranging sensor and at least one RADAR sensor. The at least one object includes an enhanced portion. The enhanced portion is substantially detectable by the at least one light detection and ranging sensor and substantially undetectable by the at least one RADAR sensor.

[0201] In an embodiment, the known characteristic data includes at least one of: information received from a remote sensor, information received from another vehicle, information received from one or more sensors at a previous time, and information determined by a pre-calibrated sensor among the one or more sensors.

[0202] In an embodiment, the known characteristic data comprises information received from a remote sensor mounted on the second vehicle.

[0203] In an embodiment, the known characteristic data comprises information received from sensors of a first type, and the sensors comprise at least one sensor of a second type different from the first type.

[0204] In an embodiment, the at least one processor executes the instructions. The at least one processor performs operations to cause the vehicle to cease operation if a calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold.

[0205] In an embodiment, the at least one processor executes the instructions. The at least one processor performs operations to notify a remote technician if a calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold.

[0206] In an embodiment, the at least one processor executes the instructions. The at least one processor performs operations to navigate the vehicle to a service location if a calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold.

[0207] In an embodiment, the at least one processor executes the instructions. The at least one processor performs operations to disable at least one sensor determined to be incorrectly calibrated if a calibration error value corresponding to the at least one sensor is greater than a calibration error threshold.

[0208] In an embodiment, a method includes: moving a vehicle and at least one object relative to each other. The vehicle includes a sensor, and the at least one object includes at least one calibration feature. While the at least one object and the vehicle are moving relative to each other, the sensor detects the at least one calibration feature of the at least one object. Each sensor includes parameters including internal parameters and external parameters. Known calibration feature data associated with the at least one calibration feature is received. Detection feature data associated with the at least one calibration feature is received from each sensor. The received known feature data is compared with the received detection feature data. For each sensor, a calibration error value is generated based on the comparison of the received known feature data with the received detection feature data. If the calibration error value corresponding to at least one sensor is greater than a first calibration error threshold, it is determined that the at least one sensor is incorrectly calibrated.

[0209] In an embodiment, a correction parameter is calculated for at least one internal parameter associated with at least one sensor determined to be incorrectly calibrated. Calculating the correction parameter is based at least in part on determining that the at least one sensor is incorrectly calibrated.

[0210] In an embodiment, said at least one internal parameter is modified based on said correction parameter.

[0211] In an embodiment, each of the sensors includes one or more monitoring devices configured to detect internal data corresponding to at least one internal parameter. The internal data associated with the at least one internal parameter is received. A correction parameter is calculated for at least one internal parameter of at least one sensor determined to be incorrectly calibrated based at least in part on the received internal data.

[0212] In an embodiment, if a calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a second calibration error threshold, it is determined that the at least one sensor is in a fault state.

[0213] In an embodiment, a correction parameter is calculated for at least one external parameter associated with at least one sensor determined to be incorrectly calibrated. Calculating the correction parameter for the at least one external parameter is based at least in part on determining that the at least one sensor is in a faulty state.

[0214] In an embodiment, the at least one object comprises three faces defining three mutually perpendicular planes. Each of the three faces comprises at least one calibration feature.

[0215] In an embodiment, the at least one object comprises a fixed road feature.

[0216] In an embodiment, the sensor comprises at least one RADAR sensor, and at least one object comprises an internal metal core detectable by the at least one RADAR sensor.

[0217] In an embodiment, the sensor includes at least one light detection and ranging sensor and at least one RADAR sensor. The at least one object includes an enhanced portion. The enhanced portion is substantially detectable by the at least one light detection and ranging sensor and substantially undetectable by the at least one RADAR sensor.

[0218] In an embodiment, the known characteristic data includes at least one of: information received from a remote sensor, information received from another vehicle, information received from one or more sensors at a previous time, and information determined by a pre-calibrated sensor among the one or more sensors.

[0219] In an embodiment, the vehicle is caused to cease operation if the calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold.

[0220] In an embodiment, if the calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold, the remote technician is notified.

[0221] In an embodiment, if the calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold, the vehicle is navigated to a service location.

[0222] In an embodiment, if a calibration error value corresponding to at least one sensor determined to be incorrectly calibrated is greater than a calibration error threshold, the at least one sensor is disabled.

[0223] In an embodiment, the known characteristic data comprises information received from a remote sensor mounted on the second vehicle.

[0224] In an embodiment, the known characteristic data comprises information received from sensors of a first type, and the sensors comprise at least one sensor of a second type different from the first type.

[0225] In the previous description, embodiments of the present invention have been described with reference to many specific details, which may vary from implementation to implementation. Therefore, the description and drawings should be regarded as illustrative, not restrictive. The only and exclusive indication of the scope of the invention, and what the applicant expects to be the scope of the invention, is the literal and equivalent scope of the claims announced from this application in the specific form of the claims of the authorization announcement, including any subsequent amendments. Any definition of terms explicitly set forth herein for being included in such claims should be based on the meaning of such terms as used in the claims. In addition, when the term "also includes" is used in the previous description or the appended claims, the phrase may be followed by additional steps or entities, or sub-steps / sub-entities of the steps or entities previously described.

Claims

1. A system for sensor calibration, comprising: at least one object configured to have a substantially reflective portion; at least one light detection and ranging sensor configured to detect a distance to at least one location associated with the substantially reflective portion of the at least one object; at least one camera sensor configured to detect a light intensity value associated with the at least one location of the substantially reflective portion of the at least one object; a computer-readable medium storing computer-executable instructions; as well as at least one processor communicatively coupled to the at least one light detection and ranging sensor and the at least one imaging sensor and configured to execute computer-executable instructions stored on the computer-readable medium, wherein when the at least one processor executes the instructions, the at least one processor performs operations to: receiving sensor data associated with the detected distance to the at least one location, generating a first predicted aggregate position associated with the substantially reflective portion based on sensor data associated with the detected distance to the at least one position, receiving sensor data associated with the detected intensity value of the at least one location, generating a second predicted aggregate position associated with the substantially reflective portion based on sensor data associated with the detected intensity value of the at least one position, and determining an alignment error value based on the first predicted aggregate position and the second predicted aggregate position, A first algorithm generates the first predicted aggregate position, and a second algorithm generates the second predicted aggregate position, the second algorithm being different from the first algorithm, wherein when the alignment error value is greater than a first alignment error threshold, at least one of the first algorithm and the second algorithm is modified, and when the alignment error value is greater than a second alignment error threshold, a calibration process of the at least one imaging sensor or the at least one light detection and ranging sensor is initiated.

2. The system according to claim 1, wherein: At least one of the first algorithm and the second algorithm is modified such that the generated first predicted aggregate position and the generated second predicted aggregate position are substantially aligned.

3. The system according to claim 1 or 2, wherein: The at least one object is further configured to have a substantially non-reflective portion.

4. The system according to claim 1 or 2, wherein: The substantially reflective portion includes at least one of a substantially black surface and a substantially white surface.

5. A method for sensor calibration, comprising: configuring at least one object to have a substantially reflective portion; detecting, with a first sensor, a distance to at least one location associated with the substantially reflective portion of the at least one object; detecting, with a second sensor, light intensity values ​​at one or more locations associated with the substantially reflective portion of the at least one object; receiving sensor data associated with the detected distance to the at least one location; generating a first predicted aggregate position associated with the substantially reflective portion based on sensor data associated with the detected distance to the at least one location; receiving sensor data associated with the detected intensity value of the at least one location; generating a second predicted aggregate position associated with the substantially reflective portion based on sensor data associated with the detected intensity value of the at least one position; as well as determining an alignment error value based on the first predicted aggregate position and the second predicted aggregate position, A first algorithm generates the first predicted aggregate position, and a second algorithm generates the second predicted aggregate position, the second algorithm being different from the first algorithm, wherein when the alignment error value is greater than a first alignment error threshold, at least one of the first algorithm and the second algorithm is modified, and when the alignment error value is greater than a second alignment error threshold, a calibration process of at least one camera sensor or at least one light detection and ranging sensor is initiated.

6. The method according to claim 5, further comprising: At least one of the first algorithm and the second algorithm is modified such that the generated first predicted aggregate position and the generated second predicted aggregate position are substantially aligned.

7. A system for sensor calibration, comprising: at least one object comprising at least one calibration feature; as well as A vehicle comprising: a plurality of sensors configured to detect the at least one calibration feature of the at least one object as the at least one object and the vehicle move relative to each other, each sensor of the plurality of sensors having a plurality of parameters, wherein the plurality of parameters includes a plurality of intrinsic parameters and a plurality of extrinsic parameters; a computer-readable medium storing computer-executable instructions; and at least one processor configured to be communicatively coupled to the plurality of sensors and to execute instructions stored on the computer-readable medium, wherein when the at least one processor executes the instructions, the at least one processor performs operations to: receiving known calibration feature data associated with the at least one calibration feature, wherein the known calibration feature data is known to be accurate; receiving, from each of the plurality of sensors, sensed feature data associated with the at least one calibration feature; comparing the received known characteristic data with the received detected characteristic data; generating, for each sensor of the plurality of sensors, a calibration error value based on a comparison of the received known feature data with the received detected feature data; and In a case where a calibration error value corresponding to at least one sensor among the plurality of sensors is greater than a first calibration error threshold, it is determined that the at least one sensor is incorrectly calibrated.

8. The system according to claim 7, wherein: When the at least one processor executes the instructions, the at least one processor further performs operations to calculate a correction parameter for at least one internal parameter associated with the at least one sensor determined to be incorrectly calibrated, wherein calculating the correction parameter is based at least in part on determining that the at least one sensor is incorrectly calibrated.

9. The system according to claim 8, wherein: When the at least one processor executes the instructions, the at least one processor further performs operations to modify the at least one internal parameter based on the correction parameter.

10. The system according to claim 8, wherein: The plurality of sensors each include at least one monitoring device configured to detect internal data corresponding to the at least one internal parameter, and wherein, when the at least one processor executes the instructions, the at least one processor further performs operations to: receiving internal data corresponding to the at least one internal parameter, and A correction parameter is calculated for the at least one internal parameter associated with the at least one sensor determined to be incorrectly calibrated, wherein the calculation of the correction parameter for the at least one internal parameter is based at least in part on the received internal data.

11. The system according to any one of claims 7 to 9, wherein: The plurality of internal parameters include at least one of the following: operating frequency, field of view, beam width, beam power, and signal-to-noise ratio.

12. The system according to any one of claims 7 to 9, wherein: When the at least one processor executes the instructions, the at least one processor further performs an operation to determine that the at least one sensor is in a fault state if a calibration error value corresponding to the at least one sensor determined to be incorrectly calibrated is greater than a second calibration error threshold.

13. The system according to claim 12, wherein: When the at least one processor executes the instructions, the at least one processor performs operations to calculate a correction parameter for at least one external parameter associated with the at least one sensor determined to be in a faulty state, wherein the calculation of the correction parameter for the at least one external parameter is based at least in part on the determination that the at least one sensor is in a faulty state.

14. The system according to any one of claims 7 to 9, wherein: The plurality of external parameters includes at least one of: a position of the sensor when the sensor is mounted on the vehicle, a distance of the sensor relative to other sensors, an angle of the sensor, a noise level caused by the environment, and an ambient brightness of the environment.

15. The system according to any one of claims 7 to 9, wherein: The at least one object includes three faces defining three mutually perpendicular planes, each of the three faces including at least one calibration feature.

16. The system according to any one of claims 7 to 9, wherein: The at least one object comprises a fixed road feature.

17. A computer program product comprising a program for causing a computer to execute the method according to claim 5 or 6.

Images