Abstract: Systems, methods, and non-transitory computer-readable media can determine sensor data captured by at least one sensor of a vehicle over a set of time intervals while navigating an environment. Three-dimensional data describing the environment over the set of time intervals can be determined from the captured sensor data. The three-dimensional data capturing a traffic motion pattern for at least one direction of travel. Image data of at least one traffic light in the environment can be determined over the set of time intervals from the captured sensor data. A state of the at least one traffic light can be predicted based at least in part on the three-dimensional data describing the environment and the image data of at least one traffic light in the environment over the set of time intervals.

Abstract: Systems, methods, and non-transitory computer-readable media can obtain information describing a static map of a geographic location, wherein the static map is determined based at least in part on a plurality of three-dimensional representations of the geographic location captured by one or more sensors of one or more vehicles. At least one training example that includes visual features and a corresponding label can be generated based on an unsupervised process for generating training examples, wherein the visual features are extracted based on the static map and at least one three-dimensional representation of the geographic location. At least one machine learning model can be trained to distinguish between static objects and non-static objects in visual data based on the at least one training example, wherein the at least one machine learning model is trained based on an unsupervised learning process.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a first label at a first position in a first image captured from a vehicle, the first label indicating that a first object is depicted in the first image at the first position, wherein the first image is a two-dimensional image. The first object is identified in a three-dimensional coordinate space representative of an environment of the vehicle based on the first position of the first label within the first image. A second label is automatically generated at a second position in a second image captured from the vehicle based on simultaneous localization and mapping (SLAM) information associated with the vehicle. The second label indicates that the first object is depicted in the second image at the second position.

Abstract: Systems, methods, and non-transitory computer-readable media can obtain and verify a hash chain. A hash chain can comprise at least a first data block and a second data block. The first data block can include a first portion of a stream of vehicle data. The second data block can include a second portion of the stream of vehicle data and a hash value of the first data block. Verification of whether the hash chain is authentic can be based at least in part on the hash value of the first data block.

Abstract: Examples disclosed herein may involve (i) identifying, in a 3D point cloud representative of a real-world environment in which a vehicle was operating during a window of time, a set of 3D data points associated with an object detected in the environment that comprises different subsets of 3D data points corresponding to different capture times within the window of time, (ii) based at least on the 3D data points, evaluating a trajectory of the object and thereby determining that the object was in motion during some portion of the window of time, (iii) in response to determining that the object was in motion, reconstructing the different subsets of 3D data points into a single, assembled 3D representation of the object, and (iv) generating a time-aggregated, 3D visualization of the environment that presents the single, assembled 3D representation of the object at one or more points along the trajectory of the object.

Abstract: In one embodiment, a method includes, determining multiple data packets for transmission. The method includes generating a log that includes a sequence identifier associated with each data packet. The method includes transmitting the data packets via a first radar antenna coupled with a vehicle and associated with the computing system. The method includes receiving an acknowledgement from a second radar antenna that identifies at least one sequence identifier associated with at least one of the data packets. The method includes identifying one or more data packets that were not received by the second radar antenna by comparing the at least one sequence identifier in the acknowledgment and the sequence identifier associated with each data packet in the log. The method includes retransmitting the one or more identified data packets that were not received by the second radar antenna via the first radar antenna.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a trajectory for a vehicle, the trajectory associated with a predicted path to be traveled by the vehicle. A power optimization plan is generated for the vehicle based on the trajectory. One or more power management settings are modified based on the power optimization plan.

Abstract: In particular embodiments, a computing system may detect a number of objects captured within an overlapping region between a first field of view associated with a first camera and a second field of view associated with a second camera. The system may determine a respective priority ranking for each of the objects. The system may select an object from the objects based on the respective priority ranking for the object. The system may determine, for the first camera, a first lighting condition associated with the first field of view. The system may determine, for the second camera, a second lighting condition associated with the second field of view. The system may determine a shared exposure time for the selected object based on the first lighting condition and the second lighting condition. The system may cause at least one image of the selected object to be captured using the shared exposure time.