Bezier-based Regression Feature Descriptor for Deformable Linear Objects

In this paper, a feature extraction approach for the deformable linear object is presented, which uses a Bezier curve to represent the original geometric shape. The proposed extraction strategy is combined with a parameterization technique, the goal is to compute the regression features from the visual-feedback RGB image, and finally obtain the efficient shape feature in the low-dimensional latent space. Existing works of literature often fail to capture the complex characteristics in a unified framework. They also struggle in scenarios where only local shape descriptors are used to guide the robot to complete the manipulation. To address these challenges, we propose a feature extraction technique using a parameterization approach to generate the regression features, which leverages the power of the Bezier curve and linear regression. The proposed extraction method effectively captures topological features and node characteristics, making it well-suited for the deformation object manipulation task. Large mount of simulations are conducted to evaluate the presented method. Our results demonstrate that the proposed method outperforms existing methods in terms of prediction accuracy, robustness, and computational efficiency. Furthermore, our approach enables the extraction of meaningful insights from the predicted links, thereby contributing to a better understanding of the shape of the deformable linear objects. Overall, this work represents a significant step forward in the use of Bezier curve for shape representation.

Sliding Mode Control for 3-D Uncalibrated and Constrained Vision-based Shape Servoing within Input Saturation

This paper designs a servo control system based on sliding mode control for the shape control of elastic objects. In order to solve the effect of non-smooth and asymmetric control saturation, a Gaussian-based continuous differentiable asymmetric saturation function is used for this goal. The proposed detection approach runs in a highly real-time manner. Meanwhile, this paper uses sliding mode control to prove that the estimation stability of the deformation Jacobian matrix and the system stability of the controller are combined, which verifies the control stability of the closed-loop system including estimation. Besides, an integral sliding mode function is designed to avoid the need for second-order derivatives of variables, which enhances the robustness of the system in actual situations. Finally, the Lyapunov theory is used to prove the consistent final boundedness of all variables of the system.

A Shape Detection Framework for Deformation Objects Using Clustering Algorithms

This paper uses clustering algorithms to introduce a shape framework for deformable objects. Until now, the shape detection of the deformable objects has faced several challenges: 1) unable to form a unified framework for multiple shapes; 2) the calculation burden as a large number of calculations; 3) the inability to solve the 3D point-cloud case. A novel shape detection framework for deformable objects is presented in this paper, which only uses the input 2D-pixel data of the objects without any artificial markers. The proposed detection approach runs in a highly real-time manner. For the definitions of the shapes of the deformable objects, three shape configurations are used to describe the outlines of the objects, i.e., centerline, contour, and surface. In addition, for the obtaining of the 3D shape, Different from the traditional 3D point cloud processing method, this article uses a one-to-one mapping method between 2D-pixel points and 3D shape points. Therefore, this guarantees a one-to-one correspondence between 2D and 3D shape points. Hence, the proposed approach can enhance the autonomous capability to detect the shape of deformable objects. Detailed experimental results are conducted within the centerline configuration to evaluate the effectiveness of the proposed shape detection framework.

Adaptive Shape-Servoing for Vision-based Robotic Manipulation with Model Estimation and Performance Regulation

This paper introduces a manipulation framework for the elastic rod, including shape representation, sensorimotor-model estimation, and shape controller. Until now, the manipulation of the elastic rod has faced several challenges: 1) shape learning from high-dimensional to low-space dimensional; 2) the modeling of robot manipulation of the elastic rod; 3) the determination of the shape controller. A novel manipulation framework for the elastic rod is presented in this paper, which only uses the input and output data of the system without any prior knowledge of the robot, camera, and object. The proposed approach runs in a model-free manner. For the approximation of the sensorimotor model, adaptive Kalman filtering (AKF) is used as the online estimation. Model-free adaptive control (MFAC) is designed according to the obtained differential model of robot-object configuration and then is combined with the performance regulation requirement to give the final format of the shape controller. Hence, the proposed approach can enhance the autonomous capability of deformation object manipulation. Detailed simulation results are conducted with a single robot manipulation to evaluate the effectiveness of the proposed manipulation framework.