Disclosed are techniques for estimating a 3D bounding box (3DBB) from a 2D bounding box (2DBB). Conventional techniques to estimate 3DBB from 2DBB rely upon classifying target vehicles within the 2DBB. When the target vehicle is misclassified, the projected bounding box from the estimated 3DBB is inaccurate. To address such issues, it is proposed to estimate the 3DBB without relying upon classifying the target vehicle.

There is provided an information processing device to realize a rich motion expression of an autonomous mobile object by easier attitude control. The information processing device includes: a motion control unit that controls a motion of an autonomous mobile object, wherein the autonomous mobile object includes a wheel that can be stored inside a main body and that can be protruded to an outside of the main body, and the motion control unit keeps a standing state by making the wheel protruded to the outside of the main body and performs driving control of the wheel and attitude control of the autonomous mobile object in movement of the autonomous mobile object, and makes the autonomous mobile object remain still in a seated state during a stop thereof by storing the wheel inside the main body is provided.

A method of integrity monitoring for visual odometry comprises capturing a first image at a first time epoch with stereo vision sensors, capturing a second image at a second time epoch, and extracting features from the images. A temporal feature matching process is performed to match the extracted features, using a feature mismatching limiting discriminator. A range, or depth, recovery process is performed to provide stereo feature matching between two images taken by the stereo vision sensors at the same time epoch, using a range error limiting discriminator. An outlier rejection process is performed using a modified RANSAC technique to limit feature moving events. Feature error magnitude and fault probabilities are characterized using overbounding Gaussian models. A state vector estimation process with integrity check is performed using solution separation to determine changes in rotation and translation between images, determine error statistics, detect faults, and compute protection level or integrity risk.

A control system that is operable to control a vehicle in an autonomous or semi-autonomous mode includes a processor that processes data captured by a plurality of exterior sensing sensors. When the control system is operating in the autonomous or semi-autonomous mode and responsive to a determination of an upcoming event that requires the system to hand over control of the vehicle to a driver before the vehicle encounters the event, the control determines (i) a total action time until the vehicle encounters the event, (ii) an estimated time for the driver to take over control and (iii) an estimated handling time for the vehicle to be controlled before the vehicle encounters the event. Responsive to the determinations, the control system (i) allows the driver to take over control of the vehicle or (ii) controls the vehicle to slow down and stop the vehicle before the vehicle encounters the event.

The assistance system for correcting vessel path comprises a receiver, a memory, at least one sensor, a processor and a display. The processor connects to the receiver, the memory and the at least one sensor, and the display is coupled to the processor. The assistance system based on the processor is able to be utilizing the preset machine-readable grid in determining the distance parameters of obstacles above water, and in which these distance parameters are corrected by using the sensor measurement data acquired by the at least one sensor. Thereby the assistance system could finish the distance measurements of abovementioned obstacles without using Lidar, and even in the case of high-undulating water surface.

An agricultural harvester includes an electromagnetic detecting and ranging module for detecting a location of an object relative to the agricultural harvester and a camera for capturing images of an area within a field of view of the electromagnetic detecting and ranging module. One or more computing devices receive first data from the electromagnetic detecting and ranging module, the first data indicating the location of the object relative to the agricultural harvester, receive image data from the camera and use the image data to determine whether the object is a receiving vehicle. If the object is a receiving vehicle, the one or more computing devices use the first data and the second data to generate graphic data defining a graphical representation illustrating the relative positions of the unload conveyor and the receiving vehicle. An electronic device including a graphical user interface presents the graphical representation on the graphical user interface.

To effectively enhance the operation efficiency of an autonomous mobile robot, an autonomous mobile robot control system includes a processor and a plurality of environmental cameras. The processor estimates a moving route of each of a plurality of moving bodies on the basis of characteristics of each of the plurality of moving bodies and sets a subset of the plurality of moving bodies whose moving routes overlap among the detected moving bodies as avoidance processing target moving bodies. The processor generates an avoidance procedure for the avoidance processing target moving bodies so the motion of the avoidance processing target moving bodies does not interfere with the motion of other avoidance target moving bodies.

An information processing system according to an embodiment of the present disclosure includes a first information processing device to be provided to a movable body and a second information processing device to be provided to a portion that differs from the movable body. The first information processing device includes a sensor portion, a generation portion, a control portion, and an integration portion. The sensor portion senses a first external environment. The generation portion uses sensor data acquired from the sensor portion to generate a first map. The control portion controls motion of a manipulator on the basis of the first map. The integration portion uses position information of inside the first external environment, with which portion the manipulator is in contact, integrates the first map and a second map acquired from the second information processing device with each other, and generates an integration map.

An automated vehicle comprising: a control unit configured to control movement of the automated vehicle to a location adjacent an estimated location of a drill hole; a scanning portion including one or more scanning devices configured to scan an area of terrain in the vicinity of the estimated location of the drill hole in order to determine an actual location of the drill hole, and to generate a point cloud representing at least a portion of the interior of the drill hole; at least one arm associated with the scanning portion, the at least one arm configured to move the scanning portion between a home position and one or more scanning positions; and an end effector associated with the at least one arm, the end effector being configured to perform one or more operations;
wherein, upon generating the point cloud, the at least one arm is configured, based on the point cloud, to position the end effector in substantial alignment with the drill hole so that the end effector can perform the one or more operations.

A method for controlling an ego agent includes capturing a two-dimensional (2D) image of an environment adjacent to the ego agent. The method also includes generating a semantically segmented image of the environment based on the 2D image. The method further includes generating a depth map of the environment based on the semantically segmented image. The method additionally includes generating a three-dimensional (3D) estimate of the environment based on the depth map. The method also includes controlling an action of the ego agent based on the identified location.