Ford Global Technolgoies, LLC

Ankit Girish Vora

Siddharth Agarwal

Helen Elizabeth Kourous-Harrigan

Sangjin Lee

Krishanth Krishnan

Jeffrey Thomas Remillard

Linjun Zhang

Codrin Cionca

Faizan Shaik

Jakob Nikolaus Hoellerbauer

Lu Xu

Devarth Parikh

Nalin Bendapudi

Peng-Yu Chen

Mostafa Parchami

Juan Enrique Castorena Martinez

Umair Ibrahim

Shenbagaraj Kannapiran

Patrick Carmody

Qi Chen

A computer that includes a processor and a memory, the memory including instructions executable by the processor to select first side stereo images of first and second pairs of stereo images acquired at first and second time steps, respectively, and mask first and second first side stereo images, determine point features in masked first and second first side stereo images and determine line features in masked first and second first side stereo images. Matching point features in the masked first and second first side stereo images can be determined using a first attention/graph neural network (attn/GNN) based on keypoints determined based on the line features. Matching line features in the masked first and second first side stereo images can be determined using a second attn/GNN based on keypoints determined based on the line features. Three-dimensional (3D) locations in a scene can be determined by determining stereo disparity based on the matched point features included in the first first side stereo image and point features determined in a first second side stereo image of first and second stereo pairs of images and a 3D stereo camera pose can be determined by determining a perspective-n-point and line algorithm on the 3D locations.

Visual odometry for operating a movable device

A system and method for operating a vehicle operating on a route that will pass within a predetermined distance of a smart infrastructure node (IX Node) includes: retrieving, prior to receiving a message from the IX Node, data about the IX Node including (a) an IX Node assist zone based on an effective broadcast range of the IX Node, (b) an IX Node field of view area that is within the IX Node assist zone and that is defined by a field of view of one or more IX Node sensors, and (c) available services from the IX Node; and selecting a vehicle application for operation on the route while in the IX Node assist zone based on the IX Node field of view area, and the available services from the IX Node, and/or a type and accuracy of one or more IX Node sensors.

Vehicle assistance in smart infrastructure node assist zone

A computer includes a processor and a memory, and the memory stores instructions executable by the processor to receive a first top-view image of an area obtained from a sensor external to a vehicle, generate a second top-view image of the area based on data from a camera of the vehicle, and determine an estimated location of the vehicle in the area based on comparing the first top-view image and the second top-view image.

Localization of a vehicle

A computer-implemented method for controlling a vehicle includes receiving, via a processor, from two or more IX control devices disposed at a two or more stationary positions having known latitudes longitudes and orientations, first sensory data identifying the position and dimensions of a feature in a mapped region. The processor generates a plurality of IX nodes based on the first sensory data received from the IX control devices, and receives LiDAR point cloud that includes LiDAR and other vehicle sensory device data such as Inertial Measurement Unit (IMU) data received from a Vehicle (AV) driving in the mapped region. The LiDAR point cloud includes a simultaneous localization and mapping (SLAM) map having second dimension information and second position information associated with the feature in the mapped region. The processor generates, without GPS and/or real-time kinematics information, an optimized High-Definition (HD) map having Absolute accuracy using batch optimization and map smoothing.

Intersection node-assisted high-definition mapping

A reference pose of an object in a coordinate system of a map of an area is determined. The reference pose is based on a three-dimensional (3D) reference model representing the object. A first pose of the object is determined as the object moves with respect to the coordinate system. The first pose is determined based on the reference pose and sensor data collected by the sensor at a first time. A second pose of the object is determined as the object continues to move with respect to the coordinate system. The second pose is determined based on the reference pose, the first pose, and sensor data collected by the sensor at a second time consecutive to the first time.

Localizing a moving object

A calibration system for a radar sensor and a method of using the system are disclosed. The method may comprise (a) receiving, from a first sensor in a vehicle, a plurality of global navigation satellite system (GNSS) parameters, wherein the plurality of GNSS parameters define a unique terrestrial position of the first sensor; (b) receiving, from a radar sensor in the vehicle, a plurality of radar parameters, wherein the plurality of radar parameters define a position of a calibration target relative to the radar sensor; (c) repeating the receiving of (a) and (b) at additional unique terrestrial positions of the first sensor; (d) using the plurality of GNSS parameters received in (a) and (c) and the plurality of radar parameters received in (b) and (c), determining corresponding positions of the calibration target; and (e) using the corresponding positions of the calibration target, determining radar calibration parameters.

Radar calibration

A system includes a processor and a memory. The memory stores instructions executable by the processor to generate, from LIDAR data, a first intensity map of an area around a vehicle, and to estimate a vehicle location based on comparing the first intensity map to a grayscale top view image.

Localization technique

A computer includes a processor and a memory storing instructions executable by the processor to receive data from a sensor specifying a plurality of points, the points including a plurality of first points that describe an object; define a boundary surrounding the first points while minimizing a volume of space that is both (i) contained by the boundary and (ii) identified as unoccupied; and actuate a component with respect to a vehicle based on the boundary.

Defining boundary for detected object

A reference pose graph can be generated from first point cloud data collected at each of a plurality of mobile nodes. A location of a stationary node can be specified based on the reference pose graph and second point cloud data collected by the stationary node.

Transportation infrastructure communication and control

Data is received from a vehicle in a stationary support structure. One or more physical attributes of an infrastructure element is determined based on the vehicle data.

In a computer mounted to a stationary support structure a vehicle and an object proximate to the support structure can be detected from data from a sensor mounted to the support structure. A risk condition can be identified based on the detected vehicle and object.

A system includes a processor and a memory. The memory stores instructions executable by the processor to select a technique, from a plurality of techniques, to identify a vehicle location by sensor and map data based on features detected in sensor data of an area in which the vehicle is located, to determine a location of the vehicle based on the selected technique; and to navigate the vehicle based at least in part on the identified location.

Localization technique selection

A computer includes a processor and a memory. The memory stores instructions executable by the processor to receive in a host vehicle, from a target vehicle via wireless vehicle-to-vehicle communications, a target vehicle first location in a global coordinate system, determine a target vehicle second location in a local coordinate system that has an origin at the host vehicle, and identify a host vehicle location in the global coordinate system based solely on (i) local coordinate system data determined in the host vehicle, including the target vehicle second location and (ii) data in the vehicle-to-vehicle communications, including the target vehicle first location.

Vehicle location identification

A system includes a light source configured to emit light in a wavelength in a visible range and a wavelength in an infrared range, and a vehicle including an infrared camera, and a computer programmed to identify an object based at least in part on an infrared image from the camera.

Street light with infrared illumination

A system includes a processor and a memory. The memory stores instructions executable by the processor to determine a plurality of waypoints for mapping an area by analyzing a received aerial image of the area, to select an optimal number of the waypoints for mapping, and to actuate the vehicle to traverse a route determined to cover the selected waypoints.

Growing community of inventors

Dearborn, MI, United States of America

Ankit Girish Vora