Abstract: A redundant processing fabric in an autonomous vehicle may include: processing, by a first processing unit of a plurality of processing units, sensor data from a first sensor of a plurality of sensors, where the plurality of processing units are coupled to the plurality of sensors via a switched fabric, wherein the plurality of processing units and plurality of sensors are included in the autonomous vehicle, wherein the sensor data corresponds to an environment external to the autonomous vehicle; determining a failure in processing the sensor data by the first processing unit; and severing, in the switched fabric, a first communications path between the first sensor and the first processing unit; and establishing, in the switched fabric, a second communications path between the first sensor and a redundant processing unit.

Abstract: Detecting stereoscopic camera desynchronization may include: receiving, at a processing unit, a first frame from a first camera and a second frame from a second camera, wherein the first camera provides first time data for when the first frame is captured by the first camera and the second camera provides second time data indicating when the second frame is captured by the second camera; associating, with the first image and independent of the first time data, a first timestamp; associating, with the second image and independent of the second time data, a second timestamp; and determining, by the processing unit, whether the first camera and the second camera are synchronized based on the first timestamp and the second timestamp.

Abstract: Systems, apparatuses, and methods for autonomously estimating the position and orientation (“pose”) of an aircraft relative to a target site are disclosed herein, including a system including a plurality of beacons arranged about the target site, wherein the plurality of beacons collectively comprise at least one electromagnetic radiation source and at least one beacon ranging radio, a sensor system coupled to the aircraft including an electromagnetic radiation sensor and a ranging radio configured to determine a range of the aircraft relative to the target site, and a processor configured to determine an estimated pose of the aircraft based on at least: (i) detected electromagnetic radiation, and (ii) time-stamped range data for the aircraft relative to the target site.

Abstract: Anticipatory vehicle electronic control unit (ECU) actuation, including: determining, based on sensor data capturing an environment around a vehicle, a predicted environmental state corresponding to a future time and comprising a transition in exterior lighting; determining, based on the predicted environmental state, an actuation state for one or more headlights of the vehicle; and configuring the one or more headlights of the vehicle according to the actuation state before the future time.

Abstract: Stereoscopic camera resynchronization in an autonomous vehicle may include: sending, to a first camera out of synchronization with a second camera, a first command to modify a frame rate of the first camera by modifying a padding data setting of the first camera; determining, independent of any determination that the first camera and second camera are resynchronized, an interval for sending, to the first camera, a second command to restore the padding data setting of the first camera; and sending, responsive to the interval occurring, the second command to the first camera.

Abstract: Extending fused multiply-add instructions, including: receiving an extended fused multiply-add (FMA) instruction comprising a first subset of bits indicating a corresponding register for each operand of a fused multiply-add (FMA) operation and a second subset of bits indicating a different register storing data describing one or more transformations applicable to one or more operands of the FMA operation; and performing, based on the extended FMA instruction.

Abstract: An autonomous electronic bicycle comprises a frame, a wheel that can be powered by a first electronic motor, and a pedal assembly connected to a pedal motor. The pedal assembly is not mechanically connected to the wheel, but the autonomous electronic bicycle simulates a mechanical connection by powering the rear wheel proportional to the user's pedaling force. The autonomous electronic bicycle uses a virtual gear ratio based on the cadence of the rider, the current incline of the bicycle, and the current speed of the bicycle. The virtual gear ratio can be a ratio between a torque of the set of pedals and a torque of the wheel.

Abstract: Automated occupant protection in stationary vehicles, including: detecting that an occupant is inside a stationary vehicle and that an operator is not in the stationary vehicle; detecting, based on one or more observations of the occupant, that one or more risk conditions have been met; and sending an alert to the operator in response to the one or more risk conditions being satisfied.

Abstract: Camera data normalization for an autonomous vehicle are described herein, including: receiving, from one or more cameras of the autonomous vehicle, camera data; applying a color normalization to the camera data; applying a spherical reprojection to the camera data; and applying, based on a registration point, a stabilization to the camera data.

Abstract: Handling input data errors in an autonomous vehicle using predictive inputs, including: determining an error in input data for a model of a plurality of models of an automation system of the autonomous vehicle; generating predicted input data for the model; and generating, based on the predicted input data, output data for the model.

Abstract: An autonomous electronic bicycle comprises a frame, a front wheel that can be powered by a first electronic motor, a rear wheel that can be powered by a second electronic motor, and handlebars that can steer the front wheel which can be controlled by a third electronic motor. Similarly, the autonomous electronic bicycle can comprise a set of sensors used for balance. The autonomous electronic bicycle can use the balance sensors to determine a current state of the autonomous electronic bicycle and drive the electronic motors to balance the autonomous electronic bicycle and achieve a target pose of the autonomous electronic bicycle. A neural network can be trained to determine one or more motor outputs of the autonomous electronic bicycle based on the target pose and the current state.

Abstract: Shared memory access in a distributed system, including: determining, in response to a memory access request, based on a time value, an entry in an access permissions table by: determining, based on a modulo of the time value and a number of entries in the access permissions table, a table index; determining, based on the table index, the entry; and determining, based on the entry, whether to allow the memory access request.

Abstract: Fault state transitions in an autonomous vehicle may include determining that a first node of a plurality of nodes has failed; determining, in response to the first node failing, a failure state; determining, based on the failure state, a configuration for the plurality of nodes excluding the first node; and applying the configuration.

Abstract: Detecting out-of-model scenarios for an autonomous vehicle including: determining, based on first sensor data from one or more sensors, an environmental state relative to the autonomous vehicle, wherein operational commands for the autonomous vehicle are based on a selected machine learning model, wherein the selected machine learning model comprises a first machine learning model; comparing the environmental state to a predicted environmental state relative to the autonomous vehicle; and determining, based on a differential between the environmental state and the predicted environmental state, whether to select a second machine learning model as the selected machine learning model.

Abstract: Scene filtering using motion estimation, including identifying, in camera data from an autonomous vehicle, based on motion relative to the autonomous vehicle, one or more pixels; filtering, from the camera data, the one or more pixels; and training, based on the filtered camera data, a neural network.

Abstract: Value-based data transmission in an autonomous vehicle, comprising: acquiring sensor data from a plurality of sensors of the autonomous vehicle, the sensor data comprising a plurality of portions; determining, for each portion of the sensor data, a value based on one or more objects identified in the sensor data; determining, based on the values for the sensor data, an upload policy; and transmitting, based on the upload policy, one or more portions of the sensor data to a server.

Abstract: Automatic disengagement of an autonomous driving mode may include receiving, from a steering torque sensor, torque sensor data indicating an amount of torque applied to a steering system of the autonomous vehicle; determining a predicted torque based on one or more motion attributes of the steering system of the autonomous vehicle; determining a differential between the predicted torque and the amount of torque; and determining, based on the differential, whether to disengage an autonomous driving mode of the autonomous vehicle.

Abstract: An autonomous electronic bicycle comprises a frame, a front wheel that can be powered by a first electronic motor, a rear wheel that can be powered by a second electronic motor, and handlebars that can steer the front wheel which can be controlled by a third electronic motor. The autonomous electronic bicycle can operate autonomously, traveling to a chosen destination. When operating autonomously the autonomous electronic bicycle can shift between a set of balance modes, including a mode used when moving and a mode used when stationary but balanced. To transition between modes, a transition sequence of control commands can be selected and used.

Abstract: A system configured to autonomously operate a vehicle within an environment is disclosed herein. The system includes a vehicle including a sensing system with a single active sensor configured to detect objects within an environment as the vehicle travels on a journey along a travel path within the environment. The system further includes a computing system communicably coupled to the vehicle. The computing system includes a memory configured to store a three-dimensional map of the environment, and a processor configured to determine an updated pose of the vehicle based on the three-dimensional map and input from the single active sensor of the vehicle. The processor is further configured to generate an updated travel path for the vehicle, wherein the updated travel path is generated based on the updated pose of the vehicle within the environment determined by the computing system.

Abstract: Device security across multiple operating system modalities, including: allocating, by a hypervisor, to a first virtual machine comprising a first operating system, based on the first modality, a first one or more access privileges to one or more resources; allocating, by the hypervisor, to a second virtual machine comprising a second operating system, based on the second modality, a second one or more access privileges to the one or more resources; and modifying, by the hypervisor, the second one or more access privileges in response to a change in an execution state of the first virtual machine.