Abstract:
The technology relates to autonomous vehicles for transporting cargo and/or people between locations. Distributed sensor arrangements may not be suitable for vehicles such as large trucks, busses or construction vehicles (100, 120). Side view mirror assemblies (300, 320) are provided that include a sensor suite of different types of sensors, including LIDAR, radar, cameras, etc. (400, 402, 404, 406). Each side assembly is rigidly secured to the vehicle by a mounting element (304, 324). The sensors within the assembly may be aligned or arranged relative to a common axis or physical point of the housing (409). This enables self -referenced calibration of all sensors in the housing. Vehicle -level calibration can also be performed between the sensors on the left and right sides of the vehicle. Each side view mirror assembly may include a conduit (234, 410) that provides one or more of power, data and cooling to the sensors in the housing.
Abstract:
In one embodiment, a system includes a first lidar sensor, which includes a first scanner configured to scan first pulses of light along a first scan pattern and a first receiver configured to detect scattered light from the first pulses of light. The system also includes a second lidar sensor, which includes a second scanner configured to scan second pulses of light along a second scan pattern and a second receiver configured to detect scattered light from the second pulses of light. The first scan pattern and the second scan pattern are at least partially overlapped. The system further includes an enclosure, where the first lidar sensor and the second lidar sensor are contained within the enclosure. The enclosure includes a window configured to transmit the first pulses of light and the second pulses of light.
Abstract:
Machine learning is applied to both 2D images from an infrared imager imaging laser reflections from an object and to the 3D depth map of the object that is generated using the 2D images and time of flight (TOF) information. In this way, the 3D depth map accuracy can be improved without increasing laser power or using high resolution imagers.
Abstract:
Apparatus for controlling a plurality of active illumination range cameras to operate in a time division multiplexed operating mode to acquire range images of scenes that the cameras image.
Abstract:
A mobile machine (10) for operating in an underground mine (20) is disclosed. The mobile machine (10) comprises a body (9) comprising a power source (15) and a cabin (28). The mobile machine (10) also comprises traction devices (13) powered by the power source (15) for propelling the machine (10) along a surface of the underground mine (20). The mobile machine also comprises earth-moving equipment (25) for at least one of (i) engaging with earth in the underground mine and (ii) hauling earth-derived payload in the underground mine, the earth-moving equipment being held by the body (9). The mobile machine also comprises an on-board ranging device (36) configured for radio frequency communication with at least one off-board ranging device (34, 37) located in the underground mine (20) to enable determination of position data for the machine (10). The position data is based on a time-based characteristic associated with transmitting a radio frequency signal to, and receiving a radio frequency signal from, the at least one off-board ranging device (34, 37). The mobile machine also comprises a mounting assembly (140) for fixing an antenna (226) to the mobile machine (10), the antenna (226) being configured for at least one of transmission and reception of the radio frequency communication. The mounting assembly (140) has a mounting base (156) for positioning the antenna (226) to be spaced from the body (9) of the mobile machine (10). The antenna (226) is mechanically protected by an electrically non-conductive material that forms a cover (141) over the antenna. The cover (141) is positioned to have a maximum height that is at or below a highest portion of the body (9) of the machine (10).
Abstract:
L'invention vise à fournir une architecture d'imageur laser à haute résolution spatiale, compatible avec une application embarquée à bord d'un véhicule, en particulier à bord d'un avion. Dans ce but, l'invention propose d'élaborer une information de télémétrie laser de type grand champ par une combinaison optique déportée adaptée. Un exemple d'équipement (1) selon l'invention embarqué dans un avion en déplacement dans un environnement susceptible de contenir des obstacles (4), en particulier d'un avion au sol, comporte un télémètre laser (11) couplé à une fibre optique (F1) d'émission d'impulsions laser (I), elle-même couplée à un système optique d'interface avec l'environnement (12) via un répartiteur optique (13) couplé à un faisceau de fibres optiques couvert sous forme d'illuminations laser (Fi). Un détecteur d'échos (14) des impulsions laser réfléchies par un obstacle (4) de l'environnement est en liaison avec une unité de traitement (15) des échos (Er), elle-même reliée à une centrale de données (16) relatives aux conditions de déplacement de l'avion et à un système d'affichage (17) des données de localisation d'obstacles (4).
Abstract:
The present disclosure describes intelligent illumination systems that use modulated light. For example, in one aspect, a method includes producing, by a light source, modulated illumination in a visible part of the spectrum, detecting three- dimensional image data based on at least a portion of the modulated light produced by the light source and reflected by a scene, and changing a characteristic of the modulated illumination based on the detected three-dimensional image data.
Abstract:
During calibration of a network of multiple horizontally scanning range finders a measurement is performed with the range finders during movement of an object through an area covered by the network. For each of the range finders data sets are identified associated with the moving object based on a static analysis. An estimated relative transformation between respective poses of the range finders is determined for reach pair of overlapping range finders, based on the identified data sets of this pair. An initial maximum likelihood configuration of poses is determined based on the estimated relative transformations of each pair. In an iterative fashion with multiple iteration steps, sets of point correspondences of the data sets are determined based on a threshold condition. A maximum likelihood configuration of poses is determined based on the sets of point correspondences and on the maximum likelihood configuration determined in the previous iteration step or, in a first iteration step, on the initial maximum likelihood configuration.
Abstract:
Various embodiments are disclosed for improved ladar transmission, including but not limited to example embodiments where closed loop feedback control is used to finely control mirror scan positions, example embodiments where range point down selection is used to improve scanning, and others.