In this study, we introduce a novel three-dimensional micro-scale robot capable of swimming in low Reynolds number. The proposed robot consists of three rotating disks linked via three perpendicular adjustable rods, actuated by three rotary and three linear motors, respectively. The robot mechanism has an important property which makes it superior to the previously designed micro swimmers. In fact, our goal is designing a micro swimmer which its controllability matrix has full rank and hence it will be capable to perform any desired maneuver in space. After introducing the mechanism and derivation of the dynamical equations of motion, we propose a control method to steer the robot to the desired position and orientation in the presence of external disturbances in the low Reynolds number flow. Simulation results confirm the successful performance of the proposed mechanism and employed control method demonstrating high maneuverability of microrobot.

Robots of next-generation physically interact with the world rather than be caged in a controlled area, and they need to make contact with the open-ended environment to perform their task. Compliant robot links offer intrinsic mechanical compliance for addressing the safety issue for physical human–robot interactions (pHRI). However, many important research questions are yet to be answered. For instance, how do system parameters, for example, mechanical compliance, motor torque, impact velocities, and so on, affect the impact force? how to formulate system impact dynamics of compliant robots, and how to size their geometric dimensions to maximize impact force reduction. In this paper, we present a parametric study of compliant link (CL) design for safe pHRI. We first present a theoretical model of the pHRI system that is comprised of robot dynamics, an impact contact model, and dummy head dynamics. After experimentally validating the theoretical model, we then systematically study the effects of CL parameters on the impact force in more detail. Specifically, we explore how the design and actuation parameters affect the impact force of pHRI system. Based on the parametric studies of the CL design, we propose a step-by-step process and a list of concrete guidelines for designing CL with safety constraints in pHRI. We further conduct a simulation case study to validate this design process and design guidelines.

Forward kinematics is essential in robot control. Its resolution remains a challenge for continuum manipulators because of their inherent flexibility. Learning-based approaches allow obtaining accurate models. However, they suffer from the explosion of the learning database that wears down the manipulator during data collection. This paper proposes an approach that combines the model and learning-based approaches. The learning database is derived from analytical equations to prevent the robot from operating for long periods. The database obtained is handled using Deep Neural Networks (DNNs). The Compact Bionic Handling robot serves as an experimental platform. The comparison with existing approaches gives satisfaction.

Human–robot interaction (HRI) is becoming more important nowadays. In this paper, a low-cost communication system for HRI is designed and implemented on the Scout robot and a robotic face. A hidden Markov model-based voice command detection system is proposed and a non-native database has been collected by Persian speakers, which contains 10 desired English commands. The experimental results confirm that the proposed system is capable to recognize the voice commands, and properly performs the task or expresses the right answer. Comparing the system with a trained system on the Julius native database shows a better true detection (about 10%).

Visual tracking is an essential building block for target tracking and capture of the underwater vehicles. On the basis of remotely autonomous control architecture, this paper has proposed an improved kernelized correlation filter (KCF) tracker and a novel fuzzy controller. The model is trained to learn an online correlation filter from a plenty of positive and negative training samples. In order to overcome the influence from occlusion, the improved KCF tracker has been designed with an added self-discrimination mechanism based on system confidence uncertainty. The novel fuzzy logic tracking controller can automatically generate and optimize fuzzy rules. Through Q-learning algorithm, the fuzzy rules are acquired through the estimating value of each state action pairs. An S surface based fitness function has been designed for the improvement of learning based particle swarm optimization. Tank and channel experiments have been carried out to verify the proposed tracker and controller through pipe tracking and target grasp on the basis of designed open frame underwater vehicle.

The soft actuator is made of superelastic material and embedded flexible material. In this paper, a kind of soft tube was designed and used to assemble two kinds of pneumatic soft actuators. The experiment and finite element analysis are used to comprehensively analyze and describe the bending, elongation, and torsion deformation of the soft actuator. The results show that the two soft actuators have the best actuation performance when the inner diameter of the soft tube is 4 mm. In addition, when the twisting pitch of the torsional actuator is 24 mm, its torsional performance is optimized. Finally, a device that can be used in the production line was assembled by utilizing those soft actuators, and some operation tasks were completed. This experiment provides some insights for the development of soft actuators with more complex motions in the future.

Localization based on visual natural landmarks is one of the state-of-the-art localization methods for automated vehicles that is, however, limited in fast motion and low-texture environments, which can lead to failure. This paper proposes an approach to solve these limitations with an extended Kalman filter (EKF) based on a state estimation algorithm that fuses information from a low-cost MEMS Inertial Measurement Unit and a Time-of-Flight camera. We demonstrate our results in an indoor environment. We show that the proposed approach does not require any global reflective landmark for localization and is fast, accurate, and easy to use with mobile robots.

A 3-RRPRR variable spherical symmetrical parallel mechanism (PM) with arc-shaped sliding pairs and no parasitic motion is presented, exhibiting two rotational and one translational (2R1T) degrees of freedom. Three limbs are symmetrically distributed between the base and end-effector; upper and lower parts of each limb are mirror symmetrical around the middle. The geometry, mobility, forward/inverse kinematics, workspace, and parasitic motion of the mechanism are analyzed, showing its ability to achieve large rotations around a continuous rotation axis. Finally, a structure synthesis strategy for variable spherical symmetrical PM is proposed, and several limb types meeting the conditions are obtained.

Interaction between a robot and its environment requires perception about the environment, which helps the robot in making a clear decision about the object type and its location. After that, the end effector will be brought to the object’s location for grasping. There are many research studies on the reaching and grasping of objects using different techniques and mechanisms for increasing accuracy and robustness during grasping and reaching tasks. Thus, this paper presents an extensive review of research directions and topics of different approaches such as sensing, learning and gripping, which have been implemented within the current five years.

Wheel slip prediction on rough terrain is crucial for secure, long-term operations of planetary exploration rovers. Although rough, unstructured terrain hampers mobility, prediction by modeling wheel–terrain interactions remains difficult owing to unclear terrain conditions and complexities of terramechanics models. This study proposes a vision-based approach with machine learning for predicting wheel slip risk by estimating the slope from 3D information and classifying terrain types from image information. It considers the slope estimation accuracy for risk prediction under sharp increases in wheel slip due to inclined ground. Experimental results obtained with a rover testbed on several terrain types validate this method.

In order to achieve high efficient self-motion for a redundant robot manipulator, a novel quadratic programming and varying-gain recurrent neural network based varying-gain neural self-motion (VGN-SM) approach is proposed and developed. With VGN-SM, the convergence errors can be adaptively and efficiently converged to zero. For comparisons, a traditional fixed-parameter neural self-motion (FPN-SM) approach is also presented. Theoretical analysis shows that the proposed VGN-SM has higher accuracy than the traditional FPN-SM. Finally, comparative experiments between VGN-SM and FPN-SM are performed on a six degrees-of-freedom robot manipulator to verify the advantages of the novel VGN-SM.