China
Africa
Explore chapters and articles related to this topic
Introduction to Visual Perception
Published in Jian Chen, Bingxi Jia, Kaixiang Zhang, Multi-View Geometry Based Visual Perception and Control of Robotic Systems, 2018
Jian Chen, Bingxi Jia, Kaixiang Zhang
Generally, reconstructed pose information are expressed with respect to the final view, which relies on the trifocal tensor among the start, current, and final images. The corresponding points and lines in the three images are usually used to estimate the trifocal tensor. For systems using perspective cameras, the three views are easy to have no correspondence due to the limited field of view. This limitation can be partially solved by using omnidirectional camera systems [14], which extend the field of view naturally. However, existing interest point detection and description algorithms suffer weak repeatability, invariance and precision with large rotational or translational displacements [81]. This means that the feature points are not consistent and subject to noise. What’s more, this problem gets worse for omnidirectional cameras because of the image distortion. In pose estimation systems, this problem is generally handled using multi-sensor fusion, such as multi-camera systems [6] and the integration of vision system, inertial measurement unit (IMU), and odometry [201]. A simple strategy is that when the image measurements are not available, the system is switched to other sensors.
A survey on vision guided robotic systems with intelligent control strategies for autonomous tasks
Published in Cogent Engineering, 2022
Abhilasha Singh, V. Kalaichelvi, R. Karthikeyan
In general, the trifocal tensor is a 3x3x3 array that calculates projective geometric relationship in all three dimensions which depends only on the pose and not on scene structure. They are more robust to outliers. Zhang et al proposed a trifocal tensor for Visual Servoing implemented for 6 DOF manipulators. The author built a trifocal tensor between current and reference views to establish a geometric relationship. Finally, a Visual controller with adaptive law was developed to compensate for the unknown distance, and stability was analysed using Lyapunov-based techniques (Zhang et al, 2019). Aranda et al. (2017) developed a 1D tensor model to estimate the angle between omnidirectional images captured at different locations. This technique provided a wide field of view with precise angles. K. K. Lee et al. (2017) used trifocal tensors to estimate the 3D motion of the camera and used two Bayesian filters for point and line measurements. B. Li et al. (2013) develop a 2D trifocal tensor-based visual stabilization controller whereas Chen Jia et al. (2017) designed a trifocal-tensor-based adaptive controller to track the desired trajectory with key video frames as input. The main advantage is that it is computationally less intensive, but the disadvantage of this method is that no closed-form solution is possible with nonlinear constraints.
Non-horizontal target measurement method based on monocular vision
Published in Systems Science & Control Engineering, 2022
Jiajin Lang, Jiafa Mao, Ronghua Liang
Since the human is binocular, the research interest of scholars on binocular vision (Hu et al., 2020; Kong et al., 2020; Liu, 2021; Mansour et al., 2019) has always been high. Ortiz et al. (2018) established the mathematical modelling of the depth error determined by the ZED camera considering a left RGB image and the depth map. And they applied the methodology to find the mathematical models of the RMS error of the Stereolabs ZED camera for all of its resolutions of operation. However, binocular vision measurements have problems such as the accuracy and synchronization between the two lenses, which directly affect the accuracy and speed of measurements. In some special engineering applications, such as pipeline measurement (Cheng et al., 2021; Haertel et al., 2015), binocular vision has the main deficiency of limited measurement range, and many scholars have carried out research on trinocular vision (Ge et al., 2021). Shao et al. (Shao & Dong, 2019) designed a trinocular vision system based on the camera with tilt-shift lens, which can enlarge the overlapping area of the trinocular vision system. And the trifocal tensor provides a stronger constraint for feature matching, which led to the method being stable and accurate for 3D reconstruction. However, in trinocular vision, different cameras acquire different data in the same scene, typically feature matching is normally according to epipolar constraint, and there is no universal algorithm to match different scenes. Measurement technologies of monocular vision with other auxiliary equipment.
Robust uncalibrated visual stabilization for nonholonomic mobile robots
Published in Advanced Robotics, 2020
For mobile robots, a common strategy used to study the geometric constraints between images is adopted to realize the visual stabilization with uncalibrated intrinsic camera parameters. The epipolar geometry defined by the current and desired images, which is isomorphic to the planar geometric setup between the initial pose and desired one of the robot system, is exploited to develop the system kinematics [14–16]. The three-dimensional distance is unknown as well as the intrinsic camera parameters. However, the epipolar geometry model is problematical with short baseline and is ill-conditioned for planar scenes. To deal with this issue, a planar projection homography-based methods are proposed in [17,18], in which the elements are organized to rebuild the system model. The chosen elements represent the relationship about the current pose of the robot system associated with the desired one. By doing so, a multi-step strategy is applied to control the translational error and angular error, respectively. Similar work is provided in [19], where the translational and angular errors could be simultaneously regulated to zero. A unified control law for visual stabilization and visual tracking is designed in [20]. Since trifocal tensor contains all the geometric relationships among the three views, which makes it independent of the observed scene. It is applied in [21,22] to overcome the drawbacks of the epipolar geometry and homography-based methods. In the aforementioned geometric constraint-based control methods, the intrinsic camera parameters are not needed to be calibrated. Unfortunately, extrinsic camera parameters are not considered. It is often assumed that the mobile robot can perform pure rotations on the spot, with the camera centered on the rotation axis, which does not correspond with the general case.