Techniques are described for determining visibility conditions of an environment in which an autonomous vehicle is operated and performing driving related operations based on the visibility conditions. An example method of adjusting driving related operations of a vehicle includes determining, by a computer located in an autonomous vehicle, a visibility related condition of an environment in which the autonomous vehicle is operating, adjusting, based at least on the visibility related condition, a set of one or more values of one or more variables associated with a driving related operation of the autonomous vehicle, and causing the autonomous vehicle to be driven to a destination by causing the driving related operation of one or more devices located in the autonomous vehicle based on at least the set of one or more values.

The disclosure relates to a method for making sensor data more robust to adversarial perturbations, wherein sensor data are obtained from at least two sensors, wherein the sensor data obtained from the at least two sensors are replaced in each case piecewise by means of quilting, wherein the piecewise replacement is carried out in such a way that the respectively replaced sensor data from different sensors are plausible relative to one another, and wherein the sensor data replaced piecewise are output.

Example data processing methods and apparatus are provided. One example method includes obtaining an image captured by an in-vehicle camera. A to-be-detected target in the image is determined. A feature region corresponding to the to-be-detected target in the image is further determined based on a location of the to-be-detected target in the image. A first parking state is determined based on the image and wheel speedometer information. A first homography matrix corresponding to the first parking state is determined from a prestored homography matrix set, where different parking states correspond to different homography matrices. Image information of the feature region is processed based on the first homography matrix to obtain a detection result.

Connected and automated vehicles (CAVs) have shown the potential to improve safety, increase road throughput, and optimize energy efficiency and emissions in several complicated traffic scenarios. This invention describes a mixed-integer programming (MIP) optimization method for global multi-vehicle decision making and motion planning of CAVs in a highly dynamic environment that consists of multiple human-driven, i.e., conventional or manual, vehicles and multiple conflict zones, such as merging points and intersections. The proposed approach ensures safety, high throughput and energy efficiency by solving a global multi-vehicle constrained optimization problem. The solution provides a feasible and optimal time schedule through road segments and conflict zones for the automated vehicles, by using information from the position, velocity, and destination of the manual vehicles, which cannot be directly controlled. Despite MIP having combinatorial complexity, the proposed formulation remains feasible for real-time implementation in the infrastructure, such as in mobile edge computers (MECs).

A vehicle sets a first determination region of a first obstacle and a second determination region, on the basis of a position of the first obstacle. The second determination region is located at a position farther than the first determination region. A reliability of the second obstacle is set to a first reliability in a case where a position of a second obstacle falls outside the first determination region and falls outside the second determination region. The reliability of the second obstacle is set to a second reliability higher than the first reliability in a case where the position of the second obstacle falls within the first determination region or falls within the second determination region. Braking is applied to the vehicle body and/or acceleration of the vehicle body is suppressed, on the basis of the reliability of the second obstacle and the position of the second obstacle.

This document describes techniques for using occlusion constraints for resolving tracks from multiple types of sensors. In aspects, an occlusion constraint is applied to an association between a radar track and vision track to indicate a probability of occlusion. In other aspects, described are techniques for a vehicle to refrain from evaluating occluded radar tracks and vision tracks collected by a perception system. The probability of occlusion is utilized for deemphasizing pairs of radar tracks and vision tracks with a high likelihood of occlusion and therefore, not useful for tracking. The disclosed techniques may provide improved perception data more closely representing multiple complex data sets for a vehicle for preventing a collision with an occluded object as the vehicle operates in an environment.

An autonomous vehicle includes an array of sensors, a processor, and a switch. The array of sensors generate sensor data related to one or more objects in an external environment of the autonomous vehicle and the processor determines an environmental context. The switch transfers the sensor data from the array of sensors to the processor, where the switch is configured to: (a) receive first sensor data from a first sensor group of the array of sensors; (b) receive second sensor data from a second sensor group of the array of sensors; (c) determine an order of transmission of the first sensor data over the second sensor data in response to the environmental context; and (d) transmit the first sensor data to the processor prior to transmitting the second sensor data based on the order of transmission.

An adaptive cruise control system includes an information acquisition unit having a main detector and a secondary detector, a control unit, and an execution unit. The main detector detects an object located ahead of the vehicle. The control unit determines whether control of the vehicle is required depending on an actual value determined by the main detector and the threshold value of a system property characterizing the driving environment. The execution unit controls the vehicle. The secondary detector is arranged such that its field of view for detecting an object located at an angle ahead of the vehicle covers the boundary of the main detector's field of view, and which is oriented outwards in relation to the forward direction of the vehicle. The secondary detector sends an indication signal to adjust the threshold value of the system property when an object is detected.

The described aspects and implementations enable fast and accurate verification of radar detection of objects in autonomous vehicle (AV) applications using combined processing of radar data and camera images. In one implementation, disclosed is a method and a system to perform the method that includes obtaining a radar data characterizing intensity of radar reflections from an environment of the AV, identifying, based on the radar data, a candidate object, obtaining a camera image depicting a region where the candidate object is located, and processing the radar data and the camera image using one or more machine-learning models to obtain a classification measure representing a likelihood that the candidate object is a real object.

A hybrid deterministic override to cloud based probabilistic advanced driver assistance systems. Under default driving conditions, an ego vehicle is controlled by a probabilistic controller in a cloud. An overall gap between the ego vehicle and a leading vehicle is divided into an emergency collision gap and a driver specified gap. The vehicle sensors monitor the overall gap. When the gap between the ego vehicle and the leading vehicle is less than or equal to the emergency collision gap, a deterministic controller of the ego vehicle overrides the cloud based probabilistic controller to control the braking and acceleration of the ego vehicle.