We propose a hierarchical cognitive navigation model (HCNM) to improve the self-learning and self-adaptive ability of mobile robots in unknown and complex environments. The HCNM model adopts the divide and conquers approach by dividing the path planning task into different levels of sub-tasks in complex environments and solves each sub-task in a smaller state subspace to decrease the state space dimensions. The HCNM model imitates animal asymptotic properties through the study of thermodynamic processes and designs a cognitive learning algorithm to achieve online optimum search strategies. We prove that the learning algorithm designed ensures that the cognitive model can converge to the optimal behavior path with probability one. Robot navigation is studied on the basis of the cognitive process. The experimental results show that the HCNM model has strong adaptability in unknown and environment, and the navigation path is clearer and the convergence time is better. Among them, the convergence time of HCNM model is 25 s, which is 86.5% lower than that of HRLM model. The HCNM model studied in this paper adopts a hierarchical structure, which reduces the learning difficulty and accelerates the learning speed in the unknown environment.

The main contribution of this paper is the design and development of the lower body of PANDORA (3D-Printed Autonomous humaNoid Developed for Open-source Research Applications), a new humanoid robotic platform implementing additive manufacturing techniques. The three joint configurations (hip, knee, and ankle) along with the major three structural parts (pelvis, thigh, and shin) of the lower body are discussed. The use of 3D printing and PLA+ material makes the robot an affordable solution for humanoid robotics research that gives a high power-to-weight ratio by significantly reducing the number of parts, as well as manufacturing and assembly time. The range of motion of the lower body of PANDORA has been investigated and is found to be comparable to a human lower body. Further, finite element analysis has been performed on the major parts of the lower body of PANDORA to check the structural integrity and to avoid catastrophic failures in the robot. The use of in-house developed actuators and robot electronics reduces the overall cost of the robot and makes PANDORA easily accessible to the research communities working in the field of humanoids. Overall, PANDORA has the potential for becoming popular between researchers and designers for investigating applications in the field of humanoid robotics, healthcare, and manufacturing, just to mention a few. The mechanical designs presented in this work are available open source to lower the knowledge barrier in developing and conducting research on bipedal robots.

To support and facilitate the rehabilitation of patients with physical limitations and aid the therapist, several robotic structures are being studied. Among the structures, the cable-driven robots stand out. The cable-driven robots are structures actuated by cables and have the advantages of being flexible and reconfigurable for each patient. The objective of this paper is to develop a theoretical model for knee flexion/extension force and moment using a cable-driven robot. The proposed model is necessary for elaborating a referential to which diagnosis can be made and the improvement of the patient evaluated. The presented theoretical model was validated through experiments with twelve sedentary and healthy volunteers. The first procedure tested ten subjects in three thigh angles for knee flexion motion; the second procedure tested two subjects in flexion and extension for the same thigh angle. The results show the validity of the model for 88.58% of the tests in an ANOVA analysis with a 99% confidence interval. The similarity of data for different gender, ages, and intrinsic factors was noted, implying that the model is representative and independent of the subject’s individuality. Differences between flexion and extension values were observed, which need to be studied in the future.

In order to make the designed ankle robotic system simpler, practical, and clinically oriented, we developed a novel $\underline{R}-2\underline{U}PS/RR$ ankle rehabilitation robot with a variety of training functions covering all the required ranges of motion of the ankle joint complex (AJC), where $U$ , $P$ , $S$ , and $R$ denote universal, prismatic, spherical, and revolute joints, respectively, and the underlined letter denotes the actuated joint. The robot was designed with three degrees of freedom (DOFs), with a series $R$ mechanism and a $2\underline{U}PS/RR$ parallel mechanism. The main advantage is that the height of the robot is very low, which is convenient for clinical use by patients. At first, the mechanism design and inverse solution of positions were introduced in detail. Then, the patient-passive exercise based on the predefined trajectory tracking and patient-active exercise based on the spring model were developed to satisfy different rehabilitation stages. Finally, experiments with healthy subjects were conducted to verify the effectiveness of the developed patient-passive and patient-active exercises of the developed ankle rehabilitation robot, with results compared with the existing ankle robotic system showing good trajectory tracking performance and interactive performance.

In the field of robot reinforcement learning (RL), the reality gap has always been a problem that restricts the robustness and generalization of algorithms. We propose Simulation Twin (SimTwin) : a deep RL framework that can help directly transfer the model from simulation to reality without any real-world training. SimTwin consists of a RL module and an adaptive correct module. We train the policy using the soft actor-critic algorithm only in a simulator with demonstration and domain randomization. In the adaptive correct module, we design and train a neural network to simulate the human error correction process using force feedback. Subsequently, we combine the above two modules through digital twin to control real-world robots, correct simulator parameters by comparing the difference between simulator and reality automatically, and then generalize the correct action through the trained policy network without additional training. We demonstrate the proposed method in an open cabinet task; the experiments show that our framework can reduce the reality gap without any real-world training.

Wearable robots, sometimes known as exoskeletons, are incredible devices for improving human strength, reducing fatigue, and restoring impaired mobility. The control of powered exoskeletons, on the other hand, is still a challenge. This necessitates the development of a technique to simulate exoskeleton–wearer interaction. This study uses a two-dimensional human skeletal model with a powered knee exoskeleton to predict the optimal lifting motion and assistive torque. For lifting motion prediction, an inverse dynamics optimization formulation is utilized. In addition, the electromechanical dynamics of the exoskeleton DC motor are modeled in the lifting optimization formulation. The design variables are human joint angle profiles and exoskeleton motor current profiles. The human joint torque square is minimized subject to physical and lifting task constraints. Then, the lifting optimization problem is solved by the gradient-based sparse nonlinear optimizer (SNOPT). Furthermore, the optimal exoskeleton torque is implemented through a two-phase control strategy to provide optimal assistance in lifting. Experimental validations of the optimal control with 6 Nm and 16 Nm maximum assistive torque are presented. Both 6 Nm and 16 Nm maximum optimal assistance of the exoskeletons reduce the mean values of vastus lateralis, biceps femoris, and latissimus dorsi muscle activations compared to the lifting without the exoskeleton. However, the mean value of the vastus medialis activation is increased by a small amount for the exoskeleton case, although its peak value is reduced. Finally, the experimental results demonstrate that the proposed lifting optimization formulation and control strategy are promising for powered knee exoskeleton for lifting tasks.

Handling and manipulating flexible porous objects is one of the main challenges in robotics for household and industrial tasks. Improving the design of grippers for flexible objects of manipulation is an important stage in the development of this topic. This article proposes a method of modeling a gripper for porous objects using the finite element method. It identifies the main parameters of the model that will affect the grasping force and the permeability of porous objects. The power characteristics of the obtained gripper model for different supply pressures, with varying porosity of the manipulated objects, are determined. The obtained characteristics are then used to find the correspondence of channel length for three textile materials with different permeable properties. An experimental study of the lifting force is conducted, and a comparison is made with the obtained modeling data for the presented samples. Additionally, using the obtained simulation data, an analysis of the pressure distribution on the surface of the porous object of manipulation is performed. As a result, it is found that the gripping device must use a design with elements to stabilize the distribution of pressure in its chamber, which will increase the stability of the gripping process.

A walking robot consisting of double Stewarts parallel legs was designed by our research team in the past time, which was mainly used for the transportation of the wounded after the disaster. In order to promote stability of control locomotion and ensure invariably horizontal state of the robot platform in the process of motion, the central pattern generator (CPG) based on particle swarm optimization (PSO) is presented to optimize the kinematic model. The purpose of optimization is to solve the hysteresis problem of displacement variation among the electric cylinders. Moreover, the dynamic model of the robot is established, which can provide mechanical basis for the feedback of control signal and make the robot move stably. The simulation results show that the displacement hysteresis problem of the electric cylinders is solved well. Meanwhile, compared with simulation results based on GA-CPG method, it is demonstrated that the robot motion planned using PSO-CPG method has better motion stability and can avoid the impact of legs landing during the transition phase of the motion cycle. The experimental results show that the platform on the robot can maintain an invariably horizontal state, and the locomotion is more stable. It verifies the feasibility of PSO-CPG model and the correctness of the dynamic model of the parallel mobile rescue robot.

The issue of implementing nonlinear model predictive control (NMPC) on mechanical systems evolving on special orthogonal group (SO(3)) is taken into consideration in the first place. Necessary conditions of optimality are extracted based on Lie group variational integrators, leading to a two-point boundary value problem (TPBVP) which is solved using sensitivity derivatives and indirect shooting methods. Fast Newton-like methods referred to as fast solvers which are commonly used to solve the TPBVP are established based on the repetition of a nonlinear process. The numerical schemes employed to alleviate the computation burden consist of eliminating some constraint-related but non-essential terms in the trend of sensitivity derivatives calculation and for solving the TPBVP equations. As another claim, assuming that a first attempt to resolve the NMPC problem is accessible, the problem subjected to some changes in its initial conditions (due to some re-planning schemes) can be resolved cost-effectively based on it. Instead of solving the whole optimization process from the scratch, the optimal control inputs and states of the system are updated based on the neighboring extremal (NE) method. For this purpose, two approaches are considered: applying NE method on the first solution that leads to a neighboring optimal solution, or assisting this latter by updating the NMPC-related optimization using exact TPBVP equations at some predefined intermediate steps. It is shown through an example that the first method is not accurate enough due to error accumulations. In contrast, the second method preserves the accuracy while reducing the computation time significantly.

This paper focuses on configuration design, dimensional synthesis, and engineering application of a novel asymmetric 2R1T parallel mechanism (PM) with zero-coupling degree. The analytical forward and inverse displacement solutions are deduced by the means of vector method. The mathematical models between Euler angles and the orientational parameters (i.e., azimuth and tilt angles) of the offset output axis are established. Using screw theory as mathematical tool, this paper worked out evaluation indices of motion/force transmissibility and presented the definitions and calculation methods of good transmission orientation workspace and good transmission orientation capacity (GTOC). Furthermore, a comparative example with respect to kinematic performance of asymmetric UPS-RPU-PU PM and planar symmetric 2UPS-PU PM is carried out, and the result demonstrates that UPS-RPU-PU significantly outperforms 2UPS-PU in terms of GTOC. The constrained optimization model is constructed to formulate optimal problem of dimensional parameters on the maximizing GTOC, which is then solved by differential evolution algorithm. Finally, an engineering case demonstrates that the optimized mechanism has a good application prospect in hydraulic support test bed.

In order to complete many complex operations and attain more general-purpose utility, robotic grasp is a necessary skill to master. As the most common essential action of robots in factory and daily life environments, robotic autonomous grasping has a wide range of application prospects and has received much attention from researchers in the past decade. However, the accurate grasp of arbitrary objects in unstructured environments is still a research challenge that has not yet been completely overcome. A complete robotic grasp system usually involves three aspects: grasp detection, grasp planning, and control subsystem. As the first step, identifying the location of the object and generating the grasp pose is the premise of successful grasp, which is conducive to planning the subsequent grasp path and the realization of the entire grasp action. Therefore, this paper conducts a literature review focusing on grasp detection technology and concludes two significant aspects: the analytic and data-driven methods. According to the previous grasp experience of the target object, this paper divides the data-driven methods into the grasp of known and unknown objects. Then it describes in detail the typical grasp detection methods and related characteristics of each classification in the grasp of unknown objects. Finally, current research status and potential research directions in this field are discussed to provide some reference for related research.

Day by day, biped robots’ usage is increasing enormously in all industrial and non-industrial applications due to their ability to move in any unstructured environment compared to wheeled robots. Keeping this in mind, worldwide, many researchers are working on various aspects of biped robots, such as gait generation, dynamic balance margin, and the design of controllers. The main aim of this review article is to discuss the main challenges encountered in the biped gait generation and design of various controllers while moving on different terrain conditions such as flat, ascending and descending slopes or stairs, avoiding obstacles/ditches, uneven terrain, and an unknown environment. As per the authors’ knowledge, no single study has been carried out in one place related to the gait generation and design of controllers for each joint of the biped robot on various terrains. This review will help researchers working in this field better understand the concepts of gait generation, dynamic balance margin, and the design of controllers while moving on various terrains. Moreover, the current article will also cover the different soft computing techniques used to tune the gains of the controllers. In this article, the authors have reviewed a vast compilation of research work on the gait generation of the biped robot on various terrains. Further, the authors have proposed taxonomies on various design issues identified while generating the gait in different aspects. The authors reviewed approximately 296 articles and discovered that all researchers attempted to generate the dynamically balanced biped gait on various terrains.

This paper focuses on the design, analysis, and multi-objective optimization of a novel 5-degrees of freedom (DOF) double-driven parallel mechanism. A novel 5-DOF parallel mechanism with two double-driven branch chains is proposed, which can serve as a machine tool. By installing two actuators on one branch chain, the proposed parallel mechanism can achieve 5-DOF of the moving platform with only three branch chains. Afterwards, analytical solution for inverse kinematics is derived. The 5$\times$5 homogeneous Jacobian matrix is obtained by transforming actuator velocities into linear velocities at three points on the moving platform. Meanwhile, the workspace, dexterity, and volume are analyzed based on the kinematic model. Ultimately, a stage-by-stage Pareto optimization method is proposed to solve the multi-objective optimization problem of this parallel mechanism. The optimization results show that the workspace, compactness, and dexterity of this mechanism can be improved efficiently.

Electronic skin (e-skin) is playing an increasingly important role in health detection, robotic teleoperation, and human-machine interaction, but most e-skins currently lack the integration of on-site signal acquisition and transmission modules. In this paper, we develop a novel flexible wearable e-skin sensing system with 11 sensing channels for robotic teleoperation. The designed sensing system is mainly composed of three components: e-skin sensor, customized flexible printed circuit (FPC), and human-machine interface. The e-skin sensor has 10 stretchable resistors distributed at the proximal and metacarpal joints of each finger respectively and 1 stretchable resistor distributed at the purlicue. The e-skin sensor can be attached to the opisthenar, and thanks to its stretchability, the sensor can detect the bent angle of the finger. The customized FPC, with WiFi module, wirelessly transmits the signal to the terminal device with human-machine interface, and we design a graphical user interface based on the Qt framework for real-time signal acquisition, storage, and display. Based on this developed e-skin system and self-developed robotic multi-fingered hand, we conduct gesture recognition and robotic multi-fingered teleoperation experiments using deep learning techniques and obtain a recognition accuracy of 91.22%. The results demonstrate that the developed e-skin sensing system has great potential in human-machine interaction.

This paper focuses on developing a novel hybrid-haptic (nHH) device with a remote center of rotation with 4 DOFs (degrees of freedom) intendant to be used as a haptic device. The new architecture is composed of two chains handling each one a part of the motions. It has the advantages of a parallel robot as high stiffness and accuracy, and the large workspace of the serial robots. The optimal synthesis of the nHH was performed using real-coded genetic algorithms. The optimization criteria and constraints were established and successively formulated and solved using a mono-objective function. A validation and comparison study were performed between the spherical parallel manipulator and the nHH. The obtained results are promising since the nHH is compared to other similar task devices, such as spherical parallel manipulator, and presents a suitable kinematic performance with a task workspace free singularity inside.

This paper presents a gait optimization method to generate the locomotion pattern for biped and discuss its stability. The main contribution of this paper is a newly proposed energy-based stability criterion, which permits the dynamic stable walking and could be straight-forwardly generalized to different locomotion scenarios and biped robots. The gait optimization problem is formulated subject to the constraints of the whole-body dynamics and kinematics. The constraints are established based on the modelling of bipedal hybrid dynamical systems. Following the whole-body modelling, the system energy is acquired and then applied to create the stability criterion. The optimization objective is also established on the system energy. The gait optimization is solved by being converted to a large-scale programming problem, where the transcription accuracy is improved via the spectral method. To further reduce the dimensionality of the large-scale problem, the whole-body dynamics is re-constructed. The generalization of the optimized gait is improved by the design of feedback control. The optimization examples demonstrate that the stability criterion naturally leads to a cyclic biped locomotion, though the periodicity was not previously imposed. Two simulation cases, level ground walking and slope walking, verify the generalization of the stability criterion and feedback control. The stability analyses are carried out by investigating the motions of centre of gravity and centre of pressure. It is revealed that if the tracked speed is above 0.3 m/s or the biped accelerates/climbs the slope, the stability criterion accomplishes the dynamic stable walking, where zero moment point criterion is not strictly complied.

The coordinated motion between the press and the feeding mechanism directly determines the production efficiency of the high-speed stamping line. In order to generate the high-performance trajectory of the feeding mechanism, this paper investigates the optimization of the trajectory with the local variable period. Based on the quintic B-spline curve and normal distribution, the smooth interpolation method of variable-time interval is proposed to generate the collision-free and energy-jerk-minimal trajectory with variable-time intervals. The advantage of the proposed method is that it can make the feeding mechanism transition smoothly between districts of variable-time and fixed-time intervals. It is beneficial to avoid re-performing the entire process of trajectory planning. ADAMS and actual experiments are used to validate the effectiveness of the proposed method. Results show that the proposed method can maintain the high performance of the initial trajectory, and there is no sharp point in the displacement-time and velocity-time curves. The investigation provides a new direction for the direct generation of local variable-period trajectories in the multi-machine coordination of the stamping line.

In an unstructured environment, the arm can perform complicated tasks with rapidity, flexibility, and robustness. It is difficult to configure multiple artificial muscles similar to an arm in the compact space of a robotic arm. When muscle tension is transferred, mechanisms like tendon-sheath/tendon-pulley may be installed in a compact space to develop musculoskeletal robots that are closer to the arm. However, handling variable frictional nonlinearity and elastic cable deformation is necessary for transmission stability. In this study, the modular artificial muscle system (MAMS), including motor cable artificial muscle and tendon sheath–pulley system (TSPS), that can be installed remotely and transmit muscle tension in narrow paths, is designed. The feed-forward multi-layer neural network (FF-MNN) approach is utilized to discuss the relationship between the measurable input tension of TSPS and the unmeasurable output tension and cable elongation. Subsequently, the lightweight musculoskeletal arm (LM-Arm) is built to verify the validity of MAMS. Through trials, the experiments of MAMS after friction compensating and the LM-Arm’s end-point 3D trajectory tracking are investigated. The results show that average errors of the active and passive muscles tension are 3.87 N and 3.51 N, respectively, under conditions of larger load and higher contraction velocity. The average muscle length error of trajectory tracking is 0.00078 m (0.72%). The suggested MAMS may successfully build a musculoskeletal robot that has similar flexibility and morphology to the arm. It can also be utilized to power various pieces of machinery, such as rescue robot, invasive surgical robots, dexterous hands, and wearable exoskeletons.