A future location of a movable object is predicted to intersect a planned path of a host vehicle. A host vehicle component is actuated to output a signal indicating to move the movable object. Then, at least one of the movable object is determined to have moved or the planned path of the host vehicle is updated. Then, the host vehicle is operated along the planned path or the updated planned path.

Smart car operations provide information or entertainment content for a person by detecting when a person is alone in a car, adjusting a speech recognizer in the car to focus capturing speech from the person's position in the car; playing content in the car as requested by the person; when the person exits the car and enters a building, transferring the speech recognizer from the car to a building speech recognizer along with a current play state of the content; and resuming playing the content on a device in the house on request without interruption.

A vehicle collision determination system includes an acceleration sensor fixedly positioned to a front end of the vehicle and configured to measure acceleration at a fixed position, a detector configured to detect whether the vehicle collides using the acceleration measured by the acceleration sensor, a metrics setting unit configured to set at least one metric using the acceleration measured by the acceleration sensor when the detector detects collision of the vehicle, and a determiner configured to determine whether an object that has collided with the vehicle is a pedestrian using the at least one metric set by the metrics setting unit.

This document describes Kurtosis based pruning for sensor-fusion systems. Kurtosis based pruning minimizes a total quantity of comparisons performed when fusing together large sets of data. Multiple candidate radar tracks may possibly align with one of multiple candidate visual tracks. For each candidate vision track, a weight or other evidence of matching is assigned to each candidate radar track. An inverse of matching errors between each candidate vision and each candidate radar track contributes to this evidence, which may be normalized to produce, for each candidate vision track, a distribution associated with all candidate radar tracks. A Kurtosis or shape of this distribution is calculated. Based on the Kurtosis values, some candidate radar tracks are selected for matching and other remaining candidate radar tracks are pruned. The Kurtosis aids in determining how many candidates to retain and how many to prune. In this way, Kurtosis based pruning can prevent combinatorial explosions due to large-scale matching.

Systems and methods for navigating a host vehicle are disclosed. In one implementation, a system includes a processor configured to receive a first image acquired by a first camera and a second image acquired by a second camera onboard the host vehicle; identify a first representation of an object in the first image and a second representation of the object in the second image; input to a first trained model at least a portion of the first image; input to a second trained model at least a portion of the second image; receive the first signature encoding determined by the first trained model and the second signature encoding determined by the second trained model; input to a third trained model the first signature encoding and the second signature encoding; and receive an indicator of a location of the object determined by the third trained model.

System, methods, and other embodiments described herein relate to determining depths of a scene from a monocular image. In one embodiment, a method includes generating depth features from sensor data according to whether the sensor data includes sparse depth data. The method includes selectively injecting the depth features into a depth model. The method includes generating a depth map from at least a monocular image using the depth model that is guided by the depth features when injected. The method includes providing the depth map as depth estimates of objects represented in the monocular image.

System, methods, and other embodiments described herein relate to determining depths of a scene from a monocular image. In one embodiment, a method includes generating depth features from depth data using a sparse auxiliary network (SAN) by i) sparsifying the depth data, ii) applying sparse residual blocks of the SAN to the depth data, and iii) densifying the depth features. The method includes generating a depth map from the depth features and a monocular image that corresponds with the depth data according to a depth model that includes the SAN. The method includes providing the depth map as depth estimates of objects represented in the monocular image.

The disclosed technology provides solutions for improving passenger drop-off functions implemented by an autonomous vehicle (AV). In some implementations, a process of the disclosed technology can include steps for collecting environmental data about an environment around an autonomous vehicle, wherein the environmental data comprises data pertaining to a roadway navigated by the autonomous vehicle, processing the environmental data to generate an area grid comprising a plurality of grid sections, and associating, based on the environmental data, one or more features with each of the plurality of grid sections. Systems and machine-readable media are also provided.

An upper-limit speed of a vehicle in automatic parking control is set, and the vehicle is caused to park automatically in a parking space at a vehicle speed equal to or lower than the set upper-limit speed. In this case, the upper-limit speed is set according to a vehicle body size of the vehicle, and the upper-limit speed is lower when the vehicle body size of the vehicle is greater than a preset size than when the vehicle body size of the vehicle is less than the preset size.

A vehicle control device includes a first prediction unit that predicts a reaching time for a vehicle to reach a cross point in front of the vehicle in a case where a pedestrian who crosses the cross point is detected based on information transmitted from a wearable device attached to the pedestrian, a second prediction unit that predicts a crossing time for the pedestrian to complete crossing the cross point based on information for specifying a walking speed acquired from the wearable device, an estimation unit that estimates whether or not the pedestrian is able to safely cross the cross point based on the predicted reaching time and the predicted crossing time, and an execution controller that executes a safe driving assistance control with respect to the vehicle in a case where the estimation unit estimates that the pedestrian is not able to safely cross the cross point.