Abstract:
In an autonomous driving system in which a plurality of autonomous driving vehicles drive under the control by occlusion control, such a situation is prevented that a vehicle is disabled to drive by a failure of an external world sensor for recognizing an obstacle and stops in an occlusion region thereby to obstruct advancement of overall succeeding vehicles to remarkably degrade the efficiency of the overall transportation work. In the autonomous driving system of the present invention, an autonomous driving vehicle which detects an obstacle or a driving road by a sensor and performs autonomous driving includes sensor state evaluation means configured to evaluate a state of performance degradation of the sensor, speed and steering angle control value setting means configured to provide limit values to the driving speed and the steering angle based on a state of performance degradation of the sensor, and movement obstacle evaluation means configured to evaluate an influence on movement of other vehicles when the vehicle stops at a position at present. When the sensor suffers from performance degradation, the vehicle stops after it drives within the set limit values to the speed and the steering angle to a point at which the vehicle does not obstruct movement of other vehicles.
Abstract:
When excavation is performed using an excavation device, a position of a bucket and a position of an excavated material are measured with high accuracy, and further, a safe excavation work is realized in the present invention. The excavation device of the present invention is characterized by including: a stereo camera; a means for recognizing a bucket from an image photographed by the camera and measuring a position of the bucket; a means for recognizing an excavated material to be separated from a ground from the image photographed by the camera and measuring a position of the excavated material; and a means for measuring a positional relationship between the bucket and the excavated material in a same screen.
Abstract:
Provided is an external environment recognition system which complements a blind spot of an on-board camera occurring due to a moving obstacle such as a preceding vehicle with a photographed image obtained from an on-road camera, and can generate a detailed map for automatic traveling. To this end, the external environment recognition system composites a photographed image from an on-board camera with a photographed image from an external camera to recognize the environment outside the host vehicle, the external environment recognition system comprising: an on-board camera which photographs an area in front of the host vehicle; an on-board communication unit which communicates with the external camera; a front moving body detection unit which detects a moving body in the front area on the basis of the photographed image from the on-board camera; a blind spot extraction unit which extracts a blind spot caused by the moving body; an image conversion unit which converts the received photographed image from the external camera into a viewpoint of the on-board camera; an image compositing unit which composites the viewpoint conversion image from the image conversion unit with the blind spot of the photographed image from the on-board camera; and a detailed map generation unit which generates a detailed map in the traveling direction of the host vehicle on the basis of the composited image from the image compositing unit.
Abstract:
The invention estimates a volume of an object inside a container without deteriorating excavation efficiency at the time of viewing the entire inside of the container with a camera. There is provided a container determination unit which determines whether an inner bottom of a bucket is within a photographing range of a stereo camera device during the work of a hydraulic excavator including the bucket and the stereo camera device; and a volume estimation unit which estimates a volume of an excavated material inside the bucket when the inner bottom of the bucket is within the photographing range of the stereo camera device.
Abstract:
The invention improves volume estimation accuracy of an object in a container using a captured image in a case where a blind spot region exists in a captured image of the object in the container. A blind spot estimation portion for estimating the blind spot region of the object in the bucket; a blind spot region shape estimation portion for estimating a shape of the object in the blind spot region; and a volume estimation portion for estimating a volume of the object in the blind spot region are included, the blind spot estimation portion estimates the blind spot region by mesh disparity data obtained from a captured image of the object in the bucket imaged by a plurality of cameras, the blind spot region shape estimation portion estimates the shape of the object in the blind spot region by the mesh disparity data, and the volume estimation portion estimates the volume of the object in the blind spot region based on the shape of the object in the blind spot region estimated by the blind spot region shape estimation portion and a shape of a bottom of the bucket.
Abstract:
An outside recognition device according to the present invention has an object to perform distance measurement of an obstacle detected with a surrounding monitoring camera at a low cost. The outside recognition device according to the present invention has a stereo camera that measures a distance in front of the vehicle, a surrounding monitoring camera that photographs a surrounding area of the vehicle, a three-dimensional measuring unit that measures a photographing range of the surrounding monitoring camera with the stereo camera by means for rotating the stereo camera, and generates three-dimensional information from the measurement result, a distance correction information generating unit that generates distance correction information corresponding to a photographing direction of surrounding monitoring camera from the three-dimensional information, and a distance measuring unit that measures a position of the obstacle detected with a surrounding monitoring camera.
Abstract:
An object recognition device includes a camera (113) and a radar (114), detects an estimated plane region estimated to be a plane from an image from the camera (113) (S202), and calculates a plane equation for the estimated plane region (S207). The object recognition device interpolates and estimates distance data within the estimated plane region based on the plane equation and the measured distance data from the radar (S210) and obtains the distance to an object by using the distance data. Accordingly, it is possible to obtain dense distance data throughout the estimated plane region and to detect the distance to small objects and distant objects.
Abstract:
An external recognition apparatus and an excavation machine using the external recognition apparatus, the external recognition apparatus including: a three-dimensional distance measurement device configured to acquire distance information in a three-dimensional space in a predetermined, region which is under a hydraulic shovel and which includes a region to be excavated by the hydraulic shovel; a plane surface estimation unit configured to estimate a plane surface in the predetermined region based on the distance information; and an excavation object region recognition unit configured to recognize the region to be excavated in the predetermined region based on the plane surface and the distance information.