SUMMARYCollision-free motion planning of a virtual arm is an intractable task in high-interference environments. In this paper, an approach for collision-free motion planning of a virtual arm based on the forward and backward reaching inverse kinematics (FABRIK) algorithm is proposed. First, a random rotation strategy and local optimum-seeking technology are introduced to improve the FABRIK algorithm in order to avoid obstacles. The improvement FABRIK algorithm is used to design the final grasping posture of a virtual arm according the target position. Then, a bidirectional rapidly exploring random tree (Bi-RRT) algorithm is adopted to explore the process postures from a given initial posture to the final grasping posture. Different from the existing method, the proposed Bi-RRT algorithm in this paper plans the motions of a virtual arm in a seven-dimensional angle space, and the final grasping posture is automatically designed using the obstacle-avoidance FABRIK algorithm instead of the manual design. Finally, the post-processing technique is introduced to remove redundant nodes from the planned motions. This procedure has resolved the problem that the Bi-RRT algorithm is a random algorithm. The experimental results show the proposed method for collision-free motion planning of a virtual arm is feasible.
Inverse kinematics (IK) has been extensively applied in the areas of robotics, computer animation, ergonomics, and gaming. Typically, IK determines the joint configurations of a robot model and achieves a desired end-effector position in robotics. Since forward and backward teaching inverse kinematics (FABRIK) is a forward and backward iterative method that finds updated joint positions by locating a point on a line instead of using angle rotations or matrices, it has the advantages of fast convergence, low computational cost, and visualizing realistic poses. However, the manipulators usually work in a complex environment. So, the kinematic chains are easy to produce the interference with their surrounding scenarios. To resolve the above mentioned problem, a two-step obstacle avoidance technology is proposed to extend the basic FABRIK in this study. The first step is a heuristic method that locates the updated linkage bar, the root joint, and the target position in a line, so that the interference can be eliminated in most cases. In the second step, multiple random rotation strategies are adopted to eliminate the interference that has not been eliminated in the first step. Experimental results have shown that the extending FABRIK has the obstacle avoidance ability.
The detection of anomalies in multivariate time-series data is becoming increasingly important in the automated and continuous monitoring of complex systems and devices due to the rapid increase in data volume and dimension. To address this challenge, we present a multivariate time-series anomaly detection model based on a dual-channel feature extraction module. The module focuses on the spatial and time features of the multivariate data using spatial short-time Fourier transform (STFT) and a graph attention network, respectively. The two features are then fused to significantly improve the model’s anomaly detection performance. In addition, the model incorporates the Huber loss function to enhance its robustness. A comparative study of the proposed model with existing state-of-the-art ones was presented to prove the effectiveness of the proposed model on three public datasets. Furthermore, by using in shield tunneling applications, we verify the effectiveness and practicality of the model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.