Deeply-learnt damped least-squares (DL-DLS) method for inverse kinematics of snake-like robots

The potential and challenges of Health 4.0 to face COVID-19 pandemic: a rapid review

The COVID-19 pandemic has generated the need to evolve health services to reduce the risk of contagion and promote a collaborative environment even remotely. Advances in Industry 4.0, including the internet of things, mobile networks, cloud computing, and artificial intelligence make Health 4.0 possible to connect patients with healthcare professionals. Hence, the focus of this work is analyzing the potentiality, and challenges of state-of-the-art Health 4.0 applications to face the COVID-19 pandemic including augmented environments, diagnosis of the virus, forecasts, medical robotics, and remote clinical services. It is concluded that Health 4.0 can be applied in the prevention of contagion, improve diagnosis, promote virtual learning environments, and offer remote services. However, there are still ethical, technical, security, and legal challenges to be addressed. Additionally, more imaging datasets for COVID-19 detection need to be made available to the scientific community. Working in the areas of opportunity will help to address the new normal. Likewise, Health 4.0 can be applied not only in the COVID-19 pandemic, but also in future global viruses and natural disasters.

Improved singular robust inverse solutions of redundant serial manipulators

To address the Jacobian matrix approximation error, which usually exists in the iterative solving process of the classic singular robust inverse method, the correction coefficient α is introduced, and the improved singular robust inverse method is the result. On this basis, the constant improved singular robust method and the intelligent improved singular robust inverse method are proposed. In addition, a new scheme, combining particle swarm optimization and artificial neural network training, is applied to obtain real-time parameters. The stability of the proposed methods is verified according to the Lyapunov stability criteria, and the effectiveness is verified in the application examples of spatial linear and curve trajectories with a seven-axis manipulator. The simulation results show that the improved singular robust inverse method has better optimization performance and stability. In the allowable range, the terminal error is smallest, and there is no lasting oscillation or large amplitude. The least singular value is largest, and the joint angular velocity is smallest, exactly as expected. The derivative of the Lyapunov function is negative definite. Comparing the two extended methods, the constant improved singular robust method performs better in terms of joint angular velocity and least singular value optimization, and the intelligent improved singular robust inverse method can achieve a smaller terminal error. There is little difference between their overall optimization effects. However, the adaptability of the real-time parameters makes the intelligent improved singular robust inverse method the first choice for kinematic control of redundant serial manipulators.

Preserving differential privacy in deep neural networks with relevance-based adaptive noise imposition

In recent years, deep learning achieves remarkable results in the field of artificial intelligence. However, the training process of deep neural networks may cause the leakage of individual privacy. Given the model and some background information of the target individual, the adversary can maliciously infer the sensitive feature of the target individual. Therefore, it is imperative to preserve the sensitive information in the training data. Differential privacy is a state-of-the-art paradigm for providing the privacy guarantee of datasets, which protects the private and sensitive information from the attack of adversaries significantly. However, the existing privacy-preserving models based on differential privacy are less than satisfactory since traditional approaches always inject the same amount of noise into parameters to preserve the sensitive information, which may impact the trade-off between the model utility and the privacy guarantee of training data. In this paper, we present a general differentially private deep neural networks learning framework based on relevance analysis, which aims to bridge the gap between private and non-private models while providing an effective privacy guarantee of sensitive information. The proposed model perturbs gradients according to the relevance between neurons in different layers and the model output. Specifically, during the process of backward propagation, more noise is added to gradients of neurons that have less relevance to the model output, and vice-versa. Experiments on five real datasets demonstrate that our mechanism not only bridges the gap between private and non-private models, but also prevents the disclosure of sensitive information effectively.

Motion and Trajectory Constraints Control Modeling for Flexible Surgical Robotic Systems

Success of the da Vinci surgical robot in the last decade has motivated the development of flexible access robots to assist clinical experts during single-port interventions of core intrabody organs. Prototypes of flexible robots have been proposed to enhance surgical tasks, such as suturing, tumor resection, and radiosurgery in human abdominal areas; nonetheless, precise constraint control models are still needed for flexible pathway navigation. In this paper, the design of a flexible snake-like robot is presented, along with the constraints model that was proposed for kinematics and dynamics control, motion trajectory planning, and obstacle avoidance during motion. Simulation of the robot and implementation of the proposed control models were done in Matlab. Several points on different circular paths were used for evaluation, and the results obtained show the model had a mean kinematic error of 0.37 ± 0.36 mm with very fast kinematics and dynamics resolution times. Furthermore, the robot’s movement was geometrically and parametrically continuous for three different trajectory cases on a circular pathway. In addition, procedures for dynamic constraint and obstacle collision detection were also proposed and validated. In the latter, a collision-avoidance scheme was kept optimal by keeping a safe distance between the robot’s links and obstacles in the workspace. Analyses of the results showed the control system was optimal in determining the necessary joint angles to reach a given target point, and motion profiles with a smooth trajectory was guaranteed, while collision with obstacles were detected a priori and avoided in close to real-time. Furthermore, the complexity and computational effort of the algorithmic models were negligibly small. Thus, the model can be used to enhance the real-time control of flexible robotic systems.