In recent years, there has been an increase in the development of medical robots to enhance interventional MRI-guided therapies and operations. Magnetic resonance imaging (MRI) surgical robots are particularly attractive due to their ability to provide excellent soft-tissue contrast during these procedures. This paper describes a novel design for a tendon-driven gripper that utilizes four shape memory alloy (SMA) spring actuators and variable stiffness joints controlled by SMA coils for use in MRI surgical robot applications. The contact force of the gripper link is determined by the mechanical properties of the SMA spring actuators (SSA) and the angle of each linkage, and the joint stiffness can be adjusted by varying the electrical current applied to the SMA coil. To enhance the efficiency of the SSAs, a new cooling system using water has been proposed and implemented. To validate the effectiveness of our proposed gripper, we conducted three types of experiments, namely, a single SSA experiment, a single SMA coil experiment, and a whole gripper experiment. The experimental results demonstrate that the proposed water-cooling system can effectively solve temperature issues of SMA, and the joint stiffness in the austenite state is higher than that in the martensite state. Moreover, our experiments show that the presented gripper is capable of grasping and holding objects of various shapes and weights.

A flexible surgical robot that can adjust its stiffness guarantees safe operation by satisfying both the high flexibility required when the robot approaches a surgical target and the more stiffness necessary for the end effector to perform surgical procedures. Therefore, this paper proposes a flexible surgical robot that consists of a central backbone, eight super-elastic wires as peripheral backbones, SMA springs, rubber tubes, and several disks. The inner diameter of the SMA springs that can be changed via their temperature creates a tightening force change to apply onto the eight backbone wires of the robot to adjust the overall stiffness of the robot. The simple structure is favorable for robot miniaturization. Frictional force and stiffness modulation experiments of the robot are performed. The results confirmed that the robot's stiffness was increased approximately 1.4 times between the martensite state and the austenite state of the SMA spring. The relationship model between the flexural rigidity of the robot and the frictional force between the SMA spring with the peripheral backbones is established. We compared the flexural rigidity values obtained from our model with those of the experimental result and confirmed that the model was valid.

Aircraft landing gear equipped with a magnetorheological (MR) damper is a semi-active system that contains nonlinear behavior, disturbances, uncertainties, and delay times that can have a huge impact on the landing’s performance. To solve this problem, this paper adopts two types of controllers, which are an intelligent controller and a model predictive controller, for a landing gear equipped with an MR damper to improve the landing gear performance considering response time in different landing cases. A model predictive controller is built based on the mathematical model of the landing gear system. An intelligent controller based on a neural network is designed and trained using a greedy bandit algorithm to improve the shock absorber efficiency at different aircraft masses and sink speeds. In this MR damper, the response time is assumed to be constant at 20 ms, which is similar to the response time of the commercial MR damper. To verify the efficiency of the proposed controllers, numerical simulations compared with a passive damper and a skyhook controller in different landing cases are executed. The major finding indicates that the suggested controller performs better in various landing scenarios than other controllers in terms of shock absorber effectiveness and adaptability.

This manuscript presents a new approach to describe aircraft landing gear systems equipped with magnetorheological (MR) dampers, integrating a reinforcement learning-based neural network control strategy. The main target of the proposed system is to improve the shock absorber efficiency in the touchdown phase, in addition to reducing the vibration due to rough ground in the taxing phase. The dynamic models of the aircraft landing system in the taxing phase with standard landing ground roughness, one-point touchdown, two-point touchdown, and third-point touchdown are built as the first step. After that, Q-learning-based reinforcement learning is developed. In order to verify the effectiveness of the controller, the co-simulations based on RECURDYN V8R4-MATLAB R2019b of the proposed system and the classical skyhook controller are executed. Based on the simulation results, the proposed controller provides better performance compared to the skyhook controller. The proposed controller provided a maximum improvement of 16% in the touchdown phase and 10% in the taxing phase compared to the skyhook controller.