Aptiv Technologies Limited

Motional Ad LLC

Delphi Technologies, Inc.

Jong Ho Lee

Junqing Wei

Junsung Kim

Wenda Xu

Gaurav Bhatia

Jarrod M Snider

Serge Lambermont

Jonathan L Wieskamp

James Esper

Zhongtao Liu

Rahul Ahluwalia

Jungsung Kim

Among other things, techniques are described for merging LiDAR information and camera information for autonomous annotation. The techniques include a vehicle that includes at least one LiDAR device configured to detect electromagnetic radiation; at least one camera configured to generate camera information of objects proximate to the vehicle; at least one computer-readable media storing computer-executable instructions; at least one processor communicatively coupled to the at least one LiDAR device and the at least one camera and a control circuit communicatively coupled to the at least one processor, wherein the control circuit is configured to operate the vehicle based on a location of the object.

Merging LiDAR information and camera information

The subject matter described in this specification is directed to a computer system and techniques for operating an autonomous vehicle (AV) based on the availability of navigation information. In one example, while the AV is in an autonomous mode, first data is obtained from a first sensor and second data is obtained from a second sensor. A location of the AV is determined based on the first data and the second data. If navigational information is not available for the determined location, an alert is provided that the AV will exit the autonomous mode.

Operation of an autonomous vehicle based on availability of navigational information

The present disclosure is directed to a computer system and techniques for automatically annotating objects in map data used for navigating an autonomous vehicle. Generally, the computer system is configured to obtain LiDAR data points for an environment around an autonomous vehicle, project the LiDAR data points onto image data, detect a target object in the image data, extract a subset of the LiDAR data points that corresponds to the detected target object, register the detected target object in map data if the extracted subset of the LiDAR data points satisfies registration criteria, and navigate the autonomous vehicle in the environment according to the map data.

Automated object annotation using fused camera/LiDAR data points

A system for operating an automated vehicle in a crowd of pedestrians includes an object-detector, optionally, a signal detector, and a controller. The object-detector detects pedestrians proximate to a host-vehicle. The signal-detector detects a signal-state displayed by a traffic-signal that displays a stop-state that indicates when the host-vehicle should stop so the pedestrians can cross in front of the host-vehicle, and displays a go-state that indicates when the pedestrians should stop passing in front of the host-vehicle so that the host-vehicle can go forward. The controller is in control of movement of the host-vehicle and in communication with the object-detector and the signal-detector. The controller operates the host-vehicle to stop the host-vehicle when the stop-state is displayed, and operates the host-vehicle to creep-forward after a wait-interval after the traffic-signal changes to the go-state when the pedestrians fail to stop passing in front of the host-vehicle.

Automated vehicle control strategy for pedestrian crowds

A driving-rule system () suitable to operate an automated includes a vehicle-detector () and a controller (). The vehicle-detector () is suitable for use on a host-vehicle (). The vehicle-detector () is used to detect movement of an other-vehicle () proximate to the host-vehicle (). The controller () is in communication with the vehicle-detector (). The controller () is configured to operate the host-vehicle () in accordance with a driving-rule (), detect an observed-deviation () of the driving-rule () by the other-vehicle (), and modify the driving-rule () based on the observed-deviation ().

Automated vehicle operating system with driving-rule modification based on observation of another vehicle

An infrastructure-device status-verification system suitable for use by an automated vehicle includes a transceiver, an object-detector, and a controller. The transceiver is suitable to install on a host-vehicle. The transceiver is used to receive an indicated-status of an infrastructure-device. The object-detector is suitable to install on the host-vehicle. The object-detector is used to determine a detected-status of the infrastructure-device. The controller is in communication with the transceiver and the object-detector. The controller determines a confirmed-status of the infrastructure-device based on the indicated-status and the detected-status. The system provides for increased confidence and security regarding information about the status of an infrastructure-device such as the traffic-signal (e.g. red, yellow, and green), thereby avoid miss-information caused by, for example, hacking or spoofing of V2I communications from the infrastructure-device, and overcome instances when the field-of-view between the object-detector and the infrastructure-devicebecomes occluded after the confirmed-status has been established or determined.

Infrastructure-device status-verification system for automated vehicles

A data-fusion system that fuses lidar-data and camera-data for an automated vehicle includes a camera, a lidar, and a controller. The camera renders an image of an object proximate to a host-vehicle. The lidar detects a distance and a direction to the object based on a reflected-signal of light reflected by the object. The controller is in communication with the camera and the lidar. The controller is configured to determine a reflectivity-characteristic of the object based on the image and the reflected-signal, and adjust a detection-characteristic of the lidar when the reflectivity-characteristic of the object makes it difficult for the lidar to detect the distance and the direction to the object.

Lidar and camera data fusion for automated vehicle

A learning system for an automated vehicle to learn local traffic customs includes a location-detector, and object-detector, and a controller. The location-detector indicates a location of a host-vehicle on a digital-map. The object-detector detects a lane-marking and other-vehicles proximate to the host-vehicle. The controller is in communication with the location-detector and the object-detector. The controller is configured to determine when an observed-behavior of the other-vehicles is not in accordance with the lane-marking present at the location, and operate the host-vehicle in accordance with the observed-behavior.

Local traffic customs learning system for automated vehicle

A system for semi-autonomous or autonomous operation of a host vehicle includes an intersection location data, at least one detector, and a controller. The intersection location data is associated with a roadway intersection. The at least one detector is configured to output a vehicle signal indicative of a location of a leading vehicle. The controller is configured to receive the vehicle signal and associate the vehicle signal relative to the intersection location data to prevent the host vehicle from entering the roadway intersection until the leading vehicle has at least traveled beyond the roadway intersection.

Traffic blocking avoidance system for an automated vehicle

A Light Detection and Ranging (LiDAR) sensor alignment system includes an imaging device, a LiDAR sensor, a mount device, and a controller. The imaging device is configured to output an image signal associated with a first scene that includes an object. The LiDAR sensor is configured to output a LiDAR signal associated with a second scene. The mount device is attached to the LiDAR sensor and adapted to align the LiDAR sensor. The controller is configured to receive the image and LiDAR signals, classify the object from the first scene, and confirm the object is properly oriented within the second scene. If the object is not properly orientated within the second scene, the controller initiates an action.

LiDAR sensor alignment system

A sensor-control system for operating an automated vehicle includes a first sensor, a second sensor, and a controller. The first sensor is used to detect objects proximate to a host-vehicle. The first sensor is characterized by a first-sensing-technology. The second sensor is used to detect objects proximate to the host-vehicle. The second sensor is characterized by a second-sensing-technology different from the first-sensing-technology. The controller is in communication with the first sensor and the second sensor. A location of an object detected by the first-sensor is used to select a field-of-view of the second-sensor.

Automated vehicle sensor control system

An object-detection system suitable for use in an automated vehicle includes a transceiver, and object-detector, and a controller. The transceiver is suitable to install on a host-vehicle. The transceiver is used to receive an indicated-location of an object proximate to the host-vehicle. The object-detector is suitable to install on the host-vehicle. The object-detector is used to detect a relative-location of the object relative to the host-vehicle. The controller is in communication with the transceiver and the object-detector. The controller determines a confirmed-location of the object relative to the host-vehicle based on the indicated-location and the relative-location. Determining the confirmed-location of the object allows for selecting from a variety of optimized algorithms to better track subsequent expected motion of the object, and reducing the risk of unnecessary or erratic action by the host-vehicle caused by hacking of the wireless-transmission or deliberate malicious content in the wireless-transmission.

Growing community of inventors

Pittsburgh, PA, United States of America

Jong Ho Lee