The application of shape memory alloy (SMA) actuated soft grippers is limited by their slow recovery speed. In order to further expand their application range, as one of the solutions to address this limitation, this paper proposes a fast actuated soft gripper based on SMA wire active heat dissipation and elastic rib combination to meet the rapid actuation and recovery. The structure primarily consists of a heat dissipation module capable of winding SMA wire and a soft structure resembling a scorpion tail with embedded supper elastic SMA wire. The geometric structure model, dynamics and SMA constitutive model and finite element model of the soft gripper are established, and the lateral deformation of soft bionic scorpion tail end is analyzed through simulations and experiments. In addition, the force in designed soft gripper tip and its ability to grasp different objects are also studied through experiments. The results show that the addition of a cooling fan increased the recovery rate by about 25%, and the force in soft bionic scorpion tail end can reach about 0.12 N. The designed soft gripper can successfully grasp objects with different softness, shape, size and weight. It lays a theoretical foundation and technical support for the development of soft grippers actuated by SMA in the future.

Vibro-impact crawling robots driven by the emerging dielectric elastomer actuators (DEAs) feature the advantages of reduced system complexity and bidirectional locomotion capability. However, due to the lack of systematic dynamic models and in-depth investigations, the fundamental locomotion principles of the robots are still unclear and therefore their applications in real-world scenarios remain hindered. In this paper, a comprehensive dynamic model of this robot is developed by considering the complex interactions between the electro-mechanical coupling, impact mechanism, and multiple nonlinear friction characteristics. By incorporating extensive modeling and experimental studies, we explain the fundamental principles that lead to the bidirectional locomotion of the robot. The actuation strategies for both forward and backward locomotion are characterized in-depth, which include the actuation frequencies, relative phases, and waveforms. Three typical contact surface cases (i.e. rigid & dry, rigid & viscous, compliant & viscous) are considered in this paper, where we report the changes in the optimal actuation strategies for bidirectional locomotion with the contact surfaces and reveal the influences of key factors of friction coefficient, viscosity, and material compliance. The key findings reported in this work can build foundations for developing a highly robust and efficient DEA-driven vibro-impact crawling robot for broad applications.

The resonant actuation of dielectric elastomer actuators (DEAs) can greatly amplify the power outputs and energy efficiencies and has facilitated numerous applications in soft robotics. In many circumstances, the DEAs are mounted on robotic bodies that are made of soft materials or compliant structures. Such compliant supports can demonstrate an equivalent stiffness and inertia comparable to the DEAs, thereby complicating the dynamics of the DEAs and threatening the performance and controllability of the soft robots by introducing additional degrees of freedom to the systems. Toward the goal of achieving reliable and controllable resonating actuation of the DEAs on soft robots, the effects of compliant supports on the dynamics of DEAs require dedicated investigations. By adopting a double cone DEA configuration, this work conducts a comprehensive study on the dynamics of a double cone DEA-compliant support system. A nonlinear dynamic model of the double cone DEA-compliant support system is developed. Together with experimental studies, the dynamics of the system in different support configurations are characterized and the influences of the key parameters in the system are clarified. The key findings of this work can potentially guide the designs of future high-performance DEA-driven soft robots.

In this paper, a multi-posture actuator (MPA) composed of VT-type bidirectional deflection fishbone-like modules (VBFM) based on shape memory alloy (SMA) wires is proposed for realizing multi-posture deformation control. Through sizing of the VT-type structure and the arrangement of VBFM in series, it is possible to realize the actuator design meeting specific deformation requirements. The thermo-mechanical coupling analysis model based on kinematics, dynamics, thermodynamics and SMA models is established to study the VBFM deflection angle and multi-posture performance of the MPA. The experimental and numerical results show that the MPA can achieve the target postures and the maximum deflection angle of the single VBFM can reach about 1.8°, and the 5-VBFM MPA can deflect a maximum angle of approximately 9° in one direction. In addition, the load actuation performance of 1-VBFM and 5-VBFM MPA is investigated, and the results show that the actuated deflection angle increases with the input electrical power within its effective range and decreases with the external resistance load. When multiple modules are combined, the deflection angle can be accumulated, and they can be used in medical robotic arms, space cleaning grab mechanisms, etc.

Purpose: The paper aims to propose a novel dual-stage shape memory alloy (SMA) actuated gripper (DAG), of which the grasp performance is improved through primary and secondary actuation. Design/methodology/approach: This paper presents a method of integrating the design of dual-stage actuation modules based on the SMA bias actuation principle to enhance the grasping shape adaptability and force modulation of a DAG. The actuation angle range and grasping performance of the DAG are investigated by thermomechanical analysis and the finite element method based numerical simulation. Findings: The results of present experiments and simulations indicate that the actuation angle scope of the DAG is about 20° under no load, which enables the grasping space occupied by an object in the DAG from 60 mm to 120 mm. The grasping force adjusted by changing the input power of the primary main actuation module and secondary fine-tuning actuation module can reach a maximum of 2 N, which is capable of grasping objects of various sizes, weights, shapes, etc. Originality/value: The contribution of this paper is to design a DAG based on SMA, and establish the solution methods for the primary main actuation module and secondary fine-tuning actuation module, respectively. It lays a foundation for the research of lightweight and intelligent robotic grippers.

In order to develop a vaccination system that have no contact with healthcare workers during highly transmissible epidemics, a portable modular vaccine injection system (PMVIS) composed of YT-type modules (YTM), winding modules (WM) and an active injector (AI) actuated by shape memory alloy (SMA) wires are proposed in this article. By series layout of WMs, YTMs and AI, the multi-posture requirements of PMVIS can be realized. Then, the thermodynamic model of SMA together with kinematic and dynamic models of structure for each module are established, and studied the output stroke of each module through numerical analysis and experiments. The results show that the designed YTM achieves a maximum output of 1.84 mm displacement or 1.9° angle under the given size parameters. After connecting 1-WM in series, the output displacement and rotation angle are enlarged to 12.12 mm and 10.8°, which are increased by 558.7% and 468.4%, respectively. PMVIS can realize multi-postures through different functional modules to play the role of position adjustment. It provides theoretical and technical support for future contactless injection systems.