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. Side view mirror assemblies are provided that include a sensor suite of different types of sensors, including LIDAR, radar, cameras, etc. Each side assembly is rigidly secured to the vehicle by a mounting element. The sensors within the assembly may be aligned or arranged relative to a common axis or physical point of the housing. 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 that provides one or more of power, data and cooling to the sensors in the housing.

A system is provided which can achieve reliability of a notification about a moving manner of a work machine for a worker regardless of the distance between the work machine and the worker. A sign image M is projected onto a peripheral region of the worker (for example, a ground surface which is present in the vicinity of the worker to the extent that the worker is capable of visually recognizing the sign image M) by an unmanned aircraft 60. The sign image M is an image which represents a moving manner of a work machine 40. Thus, regardless of the distance between the work machine 40 and the worker, reliability of a notification about the moving manner of the work machine 40 for the worker is achieved compared to a case where the sign image M is projected onto an irrelevant place to the position of the worker.

A technique for detection of an obstacle by a UAV includes arriving above a location at a first altitude by the UAV; navigating a descent flight pattern from the first altitude towards the location; acquiring aerial images of the location below the UAV with a camera system disposed onboard the UAV; and analyzing the aerial images with a machine vision system disposed onboard the UAV that is adapted to detect a presence of the obstacle in the aerial images. The descent flight pattern is selected to increase perception by the machine vision system of the obstacle.

Disclosed is an intelligent learning and adjustment system for a tennis training robot, including an image recognition system, an algorithm model, a back-end processing platform, and an optimization model. Preprocessing of incoming ball data is performed, various necessary data, such as speeds and directions of flying tennis balls, spinning and placements, are collected, various data sets are processed by using various machine learning algorithms, effective predictions and decisions are generated to facilitate the prediction of the placement and difficulty level of the incoming ball, so that a capability and level of a sparring athlete can be evaluated, the tennis training robot accordingly makes prediction and recognition, and carries out interactive feedback actions in a timely manner. The entire training process involves continuously updating of weights and bias values to make the predictions increasingly accurate, and the tennis training robot can provide an interactive intelligent training method.

Apparatuses, systems, and techniques to determine orientation of an objects in an image. In at least one embodiment, images are processed using a neural network trained to determine orientation of an object.

A method for controlling a robot device. The method includes: determining, using sensor data representing a surrounding of the robot device, whether there are one or more humans in the surrounding of the robot device; determining for each human of the one or more humans: a body pose and a velocity of the human, a predicted motion of the human for a future time period, and, using the body pose, the velocity, and the predicted motion, an occupation area occupied by the human in the future time period; generating control parameters for controlling the robot device such that a predefined distance of the robot device to the occupation area of each of the one or more humans is ensured in the future time period; and controlling the robot device in accordance with the control parameters.

A camera system includes a camera, a first light source, and a second light source. A main optical axis of the first light source intersects with a main optical axis of the second light source at an intersection point, and the intersection point is not located within a depth of field of the camera.

A crop row detection system includes a camera mounted to an agricultural machine to image a ground surface traveled by the agricultural machine to acquire time-series color images including at least a portion of the ground surface, and a processor configured or programmed to (i) perform image processing for the time-series color images, (ii) generate, from the time-series color images, an enhanced image in which a color of a crop row for detection is enhanced to provide an enhanced image, (iii) generate from the enhanced image a plan view image as viewed from above the ground surface, the plan view image being classified into first pixels having a color index value for the crop row equal to or greater than a threshold and second pixels having the color index value below the threshold, and (iv) determine positions of edge lines of the crop row based on the color index values of the first pixels.

Technology is described for a method for recognition of a target type using primitive patterns. The method can include detecting an activity signature of a target that is moving, using a sensor node. Another operation may be sampling the activity signature of the target to provide sub-samples. The sub-samples may be compared to primitives from a data store of primitives, using machine learning. In addition, the primitives are selected that are similar to the sub-samples and are joinable together to form an activity signature model. The activity signature model can be compared with activity signature templates for targets to determine the target type being captured by the sensor node.

The present invention is an autonomous mobile robot (1) that moves by being guided by a plurality of signs that have a plurality of types of sizes, are aligned along a movement path (10), and includes a first sign and a second sign, the autonomous mobile robot (1) including: an imaging unit (26); a storage unit (25) storing individual identification information of each of a plurality of signs and an individual actual size of each of the plurality of signs; and a calculation unit (27) calculating a distance (D1) to the first sign on the basis of a size of the first sign on image data captured by the imaging unit (26) and the individual actual size of the first sign corresponding to the individual identification information of the first sign.