Shanghai University

Jining University

Chongqing University

Hengyu Li

Jun Luo

Shaorong Xie

Jingyi Liu

Yueying Wang

Lei Wan

Jun Liu

Shuang Han

Yan Peng

Huayan Pu

Yuan Liu

Zhijiang Xie

Xiaomao Li

Jiajia Xie

Zhou Su

Yunyun Jiang

Disclosed is a method for constructing a bionic binocular active visual simultaneous localization and mapping (SLAM) system. The method includes: S1, constructing a binocular active visual system, peripheral vision is simulated using a panoramic camera, a panoramic value map is constructed based on the panoramic camera, and a binocular bionic eye gaze control strategy based on the panoramic value map is proposed; S2, designing a SLAM algorithm framework which is suitable for motion of the two bionic eyes, where relative pose optimization of left and right cameras based on a Perspective-n-Point algorithm and depth calculation for binocular matching points based on triangulation are adopted in a SLAM tracking thread; and S3, with the assistance of camera poses estimated by the SLAM system and sparse landmark point coordinates, training a self-supervised depth estimation network, to provide dense depth estimation for a mapping module of the SLAM system, achieving dense mapping.

Self-supervised image depth estimation method based on channel self-attention mechanism

The present disclosure provides an auto focus (AF) method and system for an electro-hydraulic (EH) lens with an adjustable focus, and an electronic device, and relates to the field of lens focusing technologies. The method mainly includes: determining a state space, an action space, and a reward function of a reinforcement learning (RL) method; constructing sample data based on the foregoing determined state space, action space, and reward function, and storing the sample data in an experience pool; and using sample data in the experience pool as training data of a deep neural network (DNN) architecture when there are M sets of the sample data in the experience pool, and obtaining a trained AF policy with reference to a deep deterministic policy gradient (DDPG) algorithm and a single hill climbing optimization (HCO) algorithm.

Auto focus (AF) method and system for electro-hydraulic (EH) lens with adjustable focus, and an electronic device

Disclosed is an active positioning method for bionic binocular vision during execution of a simultaneous localization and mapping (SLAM) task. The method includes: capturing a panoramic image using a panoramic camera and detecting key scene information; assigning values to pixels in the panoramic image to obtain a panoramic value map; projecting field-of-view areas of left and right bionic eye cameras separately onto the panoramic value map, to obtain a current binocular field-of-view projection area; calculating a mean value of the current binocular field-of-view projection area in the panoramic value map; comparing the mean value of the current binocular field-of-view projection area with a value threshold, and obtaining a binocular field-of-view projection area with a mean value higher than the value threshold by moving the binocular bionic cameras; and finally using a high-value image captured by the left and right bionic eye cameras as an input of a SLAM system.

Active positioning method for bionic binocular vision during execution of SLAM task

The present disclosure discloses an electro-hydraulic varifocal lens-based method for tracking a three-dimensional (3D) trajectory of an object by using a mobile robot. By modeling an optical imaging system, a functional relation between a focusing control current of an electro-hydraulic varifocal lens and an optimal imaging object distance is obtained. Based on this functional relation, depth information of the object in focus with respect to a mobile robot camera and an average velocity of the objective within a time interval with respect to a previous moment can be obtained. With this information, 3D coordinates and motion trajectory of the object in a camera coordinate system can be calculated. At the same time, the mobile robot locates positions and attitudes in a world coordinate system in real time, and transforms the 3D coordinates of the tracked object from the camera coordinate system to the world coordinate system.

Electro-hydraulic varifocal lens-based method for tracking three-dimensional trajectory of object by using mobile robot

The present disclosure discloses an electro-hydraulic varifocal lens-based method for tracking a 3D trajectory of a moving object. The method includes the following steps of: (1) obtaining a functional relation between a focusing control current and camera's intrinsic parameters; (2) obtaining a functional relation between focusing control currents of the electro-hydraulic varifocal lens and an optimal object distance; (3) initializing an object tracking algorithm, and taking an object tracking box as a subsequent focusing window; (4) carrying out first autofocusing, recording a focusing control current value after the autofocusing is completed, as well as a size and center point coordinates of the object tracking box; (5) calculating and recording coordinates of the object in 3D space; and (6) repeating steps (4) and (5) for the same object, and sequentially connecting the recorded coordinates of the object in 3D space into a trajectory.

Electro-hydraulic varifocal lens-based method for tracking three-dimensional trajectory of moving object

A control method and system for collaborative interception by multiple unmanned surface vessels are provided. The method includes obtaining task environment information of each unmanned surface vessel in an unmanned surface vessel group at a current moment, estimating interception point information of the intruding target at the current moment by using a Kalman filter according to the task environment information of the unmanned surface vessels at the current moment, determining process state information of each unmanned surface vessel at the current moment, inputting the process state information of each unmanned surface vessel at the current moment into a corresponding intruding target interception policy output model respectively to obtain an execution action of each unmanned surface vessel at a next moment to intercept the intruding target. The application can intercept the intruding target accurately.

Control method and system for collaborative interception by multiple unmanned surface vessels

Growing community of inventors

SHANGHAI, China

Hengyu Li