Abstract: An autonomous driving validation system may include a primary advanced autonomy system located on board an autonomous vehicle (AV) that includes sensors, computing subsystems, and a drive-by-wire system. The primary advanced autonomy system may be configured to process first computations relating to a driving operation of the AV. The autonomous driving validation system may also include an off-board autonomous cloud computing system that includes cloud computing subsystems. The off-board autonomous cloud computing system may be configured to process second computations that are similar to the first computations processed by the primary advanced autonomy system.

Abstract: A method may include obtaining input information relating to an environment in which an autonomous vehicle (AV) operates. The input information may describe a state of the AV, an operation of the AV within the environment, a property of the environment, or an object included in the environment. The method may include setting up a traffic rule profile for the AV that specifies societal traffic practices corresponding to a location of the environment. The method may include identifying a first traffic rule relevant to the AV based on the obtained input information and determining a first decision corresponding to the traffic rule profile and the first traffic rule. The method may include sending an instruction to a control system of the AV, the instruction describing a given operation of the AV responsive to the traffic rule profile and the first traffic rule according to the first decision.

Abstract: A method may include obtaining input information relating to an environment in which an autonomous vehicle (AV) operates in which the input information describes at least one of: a state of the AV, an operation of the AV within the environment, a property of the environment, or an object included in the environment. The method may include identifying a region of interest that represents a section of the environment based on the input information and identifying a portion of the environment based on the identified region of interest. The portion of the environment may include an object that affects operation of the AV. The method may include determining a first decision that relates to the object and sending an instruction to a control system of the AV describing a given operation of the AV responsive to the object according to the first decision.

Abstract: A method may include identifying computing systems corresponding to an autonomous vehicle (AV) in which each computing system is configured to perform at least one operation relating to driving the AV. The method may include determining whether the computing systems use a first respective real-world parameter as an input to any of the operations or generate a second respective real-world parameter as an output of the operations. A respective operation corresponding to a respective computing system may be designated for accelerated performance based on the determination, and the operations of the computing systems may be performed. A first operation not designated for accelerated performance may include first computations corresponding to the first operation after waiting for a synchronization delay period, and a second operation designated for accelerated performance may include second computations corresponding to the second operation and performed without waiting for the synchronization delay period.

Abstract: A method may include obtaining, by an autonomous driving system, input information relating to an autonomous vehicle (AV). The method may include determining, by the autonomous driving system, one or more driving signals that describe operations of the AV. The method may include instructing the AV to move according to the driving signals and simulating, by a virtual driving system, movement of the AV based on the driving signals, wherein the simulating the movement of the AV occurs concurrently with the AV moving according to the driving signals as instructed.

Abstract: A method may include obtaining sensor data relating to an autonomous vehicle (AV) and a total measurable world around the AV. The method may include identifying an operating environment of the AV and determining a projected computational load for computing subsystems that facilitate a driving operation performable by the AV corresponding to the identified environment. The method may include off-loading first computing subsystems of the computing subsystems in which computations of the first computing subsystems may be processed by an off-board cloud computing system and processing computations associated with second computing subsystems of the computing subsystems by an on-board computing system.

Abstract: A method may include obtaining one or more inputs in which each of the inputs describes at least one of: a state of an autonomous vehicle (AV) or a state of an object; and identifying a prediction context of the AV based on the inputs. The method may also include determining a relevancy of each object of a plurality of objects to the AV in relation to the prediction context; and outputting a set of relevant objects based on the relevancy determination for each of the plurality of objects. Another method may include obtaining a set of objects designated as relevant to operation of an AV; selecting a trajectory prediction approach for a given object based on context of the AV and characteristics of the given object; predicting a trajectory of the given object using the selected trajectory prediction approach; and outputting the given object and the predicted trajectory.

Abstract: A method may include obtaining sensor data from one or more LiDAR units and determining a point-cloud corresponding to the sensor data obtained from each respective LiDAR unit. The method may include aggregating the point-clouds as an aggregated point-cloud. A number of data points included in the aggregated point-cloud may be decreased by filtering out one or more of the data points according to one or more heuristic rules to generate a reduced point-cloud. The method may include determining an operational granularity level for the reduced point-cloud. An array of existence-based objects may be generated based on the reduced point-cloud and the operational granularity level.

Abstract: A method may include obtaining input information that describes a driving operation of a vehicle and obtaining a rule that indicates an approved driving operation of the vehicle. The method may include parsing the rule using a domain-specific language to generate rule conditions in which the domain-specific language is a programming language that is specifically designed for analyzing driving operations of vehicles. The method may include representing the input information as observations relating to the vehicle in which each of the observations is comparable to one or more of the rules. The method may include comparing the observations to one or more respective comparable rule conditions and generating a grading summary that evaluates how well the observations satisfy the respective comparable rule conditions based on the comparison. A future driving operation of the vehicle may be adjusted based on the grading summary.

Abstract: A modular sensor system may include railings that connectively couple with one another. The modular sensor system may include a first compartment body that couples to the railings. The modular sensor system may include sensors positioned in an interior region or on an exterior surface of the first compartment body. In some embodiments, the railings may be connectively coupled to define an interior space that includes a shape corresponding to at least a partial perimeter shape of an industrial machine with the interior space of the modular sensor system interfacing with the industrial machine. In some embodiments, a second compartment body may be coupled to the railings. In some embodiments, the modular sensor system may include a surface panel that pivots along an axis of rotation from a first position perpendicular to the railings to a second position parallel to the railings and covering the interior space.

Abstract: Vehicle sensor systems include modular sensor kits having one or more pods (e.g., sensor roof pods) and/or one or more bumpers (e.g., sensor bumpers). The sensor roof pods are configured to couple to a vehicle. A sensor roof pod may be positioned atop a vehicle proximate a front of the vehicle, proximate a back of the vehicle, or at any position along a top side of the vehicle being coupled, for example, using a mounting shim or a tripod. The sensor roof pods can include sensors (e.g., LIDAR sensors, cameras, ultrasonic sensors, etc.), processing units, control systems (e.g., temperature and/or environmental control systems), and communication devices (e.g., networking and/or wireless devices).

Abstract: A method may include obtaining one or more inputs in which each of the inputs describes at least one of: a state of an autonomous vehicle (AV) or a state of an object; and identifying a prediction context of the AV based on the inputs. The method may also include determining a relevancy of each object of a plurality of objects to the AV in relation to the prediction context; and outputting a set of relevant objects based on the relevancy determination for each of the plurality of objects. Another method may include obtaining a set of objects designated as relevant to operation of an AV; selecting a trajectory prediction approach for a given object based on context of the AV and characteristics of the given object; predicting a trajectory of the given object using the selected trajectory prediction approach; and outputting the given object and the predicted trajectory.

Abstract: Virtual bumpers for autonomous vehicles improve effectiveness and safety as such vehicles are operated. One or more sensor systems having a Lidar sensor and a camera sensor determine proximity of objects around the vehicle and facilitate identification of the environment around the vehicle. The sensor systems are placed at various locations around the vehicle. The vehicle identifies an object and one or more properties of the identified object using the sensor systems. Based on the identified object and the properties of the object, a virtual bumper may be created for that object. For example, if the object is identified as another vehicle moving with a certain velocity, the vehicle may determine a minimum space to avoid the other vehicle, either by changing direction or reducing the velocity of the vehicle, with the minimum space constituting a virtual bumper.

Abstract: A method may include obtaining first sensor data captured by a first sensor system and second sensor data captured by a second sensor system of a different type from the first sensor system. The method may include detecting a first object included in the first sensor data and a second object included in the second sensor data. The method may include assigning a first label to the first object and a second label to the second object after comparing the first and the second sensor data. The first and second labels may indicate degrees to which the first and the second objects match. Responsive to the first and second labels indicating that the first and the second objects match, the method may include designating a matched object representative of the first object and the second object and sending the matched object to a downstream computing system of an autonomous vehicle.

Abstract: A method may include obtaining input information that describes a driving operation of a vehicle and obtaining a rule that indicates an approved driving operation of the vehicle. The method may include parsing the rule using a domain-specific language to generate rule conditions in which the domain-specific language is a programming language that is specifically designed for analyzing driving operations of vehicles. The method may include representing the input information as observations relating to the vehicle in which each of the observations is comparable to one or more of the rules. The method may include comparing the observations to one or more respective comparable rule conditions and generating a grading summary that evaluates how well the observations satisfy the respective comparable rule conditions based on the comparison. A future driving operation of the vehicle may be adjusted based on the grading summary.

Abstract: A method may include identifying computing systems corresponding to an autonomous vehicle (AV) in which each computing system is configured to perform at least one operation relating to driving the AV. The method may include determining whether the computing systems use a first respective real-world parameter as an input to any of the operations or generate a second respective real-world parameter as an output of the operations. A respective operation corresponding to a respective computing system may be designated for accelerated performance based on the determination, and the operations of the computing systems may be performed. A first operation not designated for accelerated performance may include first computations corresponding to the first operation after waiting for a synchronization delay period, and a second operation designated for accelerated performance may include second computations corresponding to the second operation and performed without waiting for the synchronization delay period.

Abstract: A method may include obtaining, by an autonomous driving system, input information relating to an autonomous vehicle (AV). The method may include determining, by the autonomous driving system, one or more driving signals that describe operations of the AV. The method may include instructing the AV to move according to the driving signals and simulating, by a virtual driving system, movement of the AV based on the driving signals, wherein the simulating the movement of the AV occurs concurrently with the AV moving according to the driving signals as instructed.

Abstract: A method may include determining an alignment time based on a zero-crossing point corresponding to a LiDAR sensor and a horizontal field of view corresponding to an image-capturing sensor. The method may include determining a delay timing for initiating image capturing by the image-capturing sensor in which the delay timing is based on at least one of: the alignment time, a packet capture timing corresponding to the LiDAR sensor, and an average frame exposure duration corresponding to the image-capturing sensor. The method may include initiating data capture by the LiDAR sensor, and after the initiating of data capture by the LiDAR sensor and after the delay timing has elapsed, initiating data capture by the image-capturing sensor.

Abstract: A method may include obtaining localization information and control signal information relating to an autonomous vehicle (AV). The method may include computing first kinematic characteristics of the AV based on the localization information and the control signal information and generating a first vehicle model based on the first kinematic characteristics. The method may include determining one or more first control commands that provide instructions for actuating a driving operation of the AV. The first control commands may be determined based on the first vehicle model. The method may include simulating first movement of the AV according to the first control commands and the first vehicle model.

Abstract: An autonomous driving validation system may include a primary advanced autonomy system located on board an autonomous vehicle (AV) that includes sensors, computing subsystems, and a drive-by-wire system. The primary advanced autonomy system may be configured to process first computations relating to a driving operation of the AV. The autonomous driving validation system may also include an off-board autonomous cloud computing system that includes cloud computing subsystems. The off-board autonomous cloud computing system may be configured to process second computations that are similar to the first computations processed by the primary advanced autonomy system.