With the rise of deep reinforcement learning (RL) methods, many complex robotic manipulation tasks are being solved. However, harnessing the full power of deep learning requires large datasets. Online RL does not suit itself readily into this paradigm due to costly and time-consuming agent-environment interaction. Therefore, many offline RL algorithms have recently been proposed to learn robotic tasks. But mainly, all such methods focus on a single-task or multitask learning, which requires retraining whenever we need to learn a new task. Continuously learning tasks without forgetting previous knowledge combined with the power of offline deep RL would allow us to scale the number of tasks by adding them one after another. This paper investigates the effectiveness of regularisation-based methods like synaptic intelligence for sequentially learning image-based robotic manipulation tasks in an offline-RL setup. We evaluate the performance of this combined framework against common challenges of sequential learning: catastrophic forgetting and forward knowledge transfer. We performed experiments with different task combinations to analyse the effect of task ordering. We also investigated the effect of the number of object configurations and the density of robot trajectories. We found that learning tasks sequentially helps in the retention of knowledge from previous tasks, thereby reducing the time required to learn a new task. Regularisation-based approaches for continuous learning, like the synaptic intelligence method, help mitigate catastrophic forgetting but have shown only limited transfer of knowledge from previous tasks.

This paper presents a compliant variable admittance adaptive fixed-time sliding mode control (SMC) algorithm for trajectory tracking of robotic manipulators. Specifically, a compliant variable admittance algorithm and an adaptive fixed-time SMC algorithm are combined to construct a double-loop control structure. In the outer loop, the variable admittance algorithm is developed to adjust admittance parameters during a collision to minimize the collision time, which gives the robot compliance property and reduce the rigid collision influence. Then, by employing the Lyapunov theory and the fixed-time stability theory, a new nonsingular sliding mode manifold is proposed and an adaptive fixed-time SMC algorithm is presented in the inner loop. More precisely, this approach enables rapid convergence, enhanced steady-state tracking precision, and a settling time that is independent of system initial states. As a result, the effectiveness and improved performance of the proposed algorithm are demonstrated through extensive simulations and experimental results.

In the process of trajectory optimization for robot manipulator, the path that is generated may deviate from the intended path because of the adjustment of trajectory parameters, if there is limitation of end-effector path in Cartesian space for specific tasks, this phenomenon is dangerous. This paper proposes a methodology that is based on the Pareto front to address this issue, and the methodology takes into account both the multi-objective optimization of robotic arm and the quality of end-effector path. Based on dung beetle optimizer, this research proposes improved non-dominated sorting dung beetle optimizer. This paper interpolates manipulator trajectory with quintic B-spline curves, achieves multi-objective trajectory optimization that simultaneously optimizes traveling time, energy consumption, and mean jerk, proposes a trajectory selection strategy that is based on Pareto solution set by introducing the concept of Fréchet distance, and the strategy enables the end-effector to approach the desired path in Cartesian space. Simulation and experimental results validate the effectiveness and practicability of the proposed methodology on the Sawyer robot manipulator.

In this paper, an $\textrm{H}_{{\infty }}$ dynamic output feedback controller is experimentally implemented for the position regulation of a fully actuated tilted-rotor octocopter unmanned aerial vehicle (UAV) to improve wind disturbance rejection during station-keeping. To apply the lateral forces, besides the standard tilt-to-translate (attitude-thrust) movement, tilted-rotor UAVs can generate vectored (horizontal) thrust. Vectored-thrust is high-bandwidth but saturation-constrained, while attitude-thrust generates larger forces with lower bandwidth. For the first time, this paper emphasizes the frequency-dependent allocation of weighting matrices in $\textrm{H}_{{\infty }}$ control design based on the physical capabilities of the fully actuated UAV (vectored-thrust and attitude-thrust). A dynamic model of the tilted-rotor octocopter, including aerodynamic effects and rotor dynamics, is presented to design the controller. The proposed $\textrm{H}_{{\infty }}$ controller solves the frequency-dependent actuator allocation problem by augmenting the dynamic model with weighting transfer functions. This novel frequency-dependent allocation utilizes the attitude-thrust for low-frequency disturbances and vectored-thrust for high-frequency disturbances, which exploits the maximum potential of the fully actuated UAV. Several wind tunnel experiments are conducted to validate the model and wind disturbance rejection performance, and the results are compared to the baseline PX4 Autopilot controller on both the tilted-rotor and a planar octocopter. The $\textrm{H}_{{\infty }}$controller is shown to reduce station-keeping error by up to 50% for an actuator usage 25% higher in free-flight tests.

The manufacturing of the X-shaped tip of prestressed centrifugal concrete piles is nowadays done half automatically by combining the manual worker and the automatic welding robot. To make this welding process full automatically, the welding seam tracking algorithm is considered. There are many types of sensors that can be used to detect the welding seam such as vision sensor, laser vision sensor, arc sensor, or touch sensor. Each type of sensor has its advantages and disadvantages. In this paper, an algorithm for welding seam tracking using laser distance sensor is proposed. Firstly, the fundamental mathematics theory of the algorithm is presented. Next, the positioning table system supports the procedure is designed and manufactured. The object of this research is the fillet joint because of the characteristics of the X-shaped tip of the concrete piles. This paper proposes a new method to determine the welding trajectory of the tip using laser distance sensor. After that, the experimental results are received to verify the proposed idea. Finally, the improved proposal of the algorithm is considered to increase the accuracy of the suggested algorithm.

In this paper, an online adaptive super twisting sliding mode controller is proposed for a non-linear system. The adaptive controller has been designed in order to deal with the unknown dynamic uncertainties and give the best trajectory tracking. The adaptation is based on an optimal Particle Swarm Optimization (PSO) algorithm whose goal is online tuning the parameters through focusing on decreasing the objective function. The novelty of this study is online handling parameters setting in the conventional super twisting algorithm, bypass heavy offline calculation, and also avoid the instability and abrupt changing of the controller’s parameters for better actuators lifetime. This novel approach has been applied on an upper limb exoskeleton robot for arm rehabilitation. Despite the changes of the dynamic model of the system which defers from one patient to another due to the direct interactions between the wearer and the exoskeleton, this control technique preserves its robustness with respect to bounded external disturbances. The effectiveness of the proposed adaptive controller has been proved in simulation and then in real-time experiment with two human subjects. A comparison between the proposed approach and classic super twisting algorithm has been conducted. The obtained results show the performance and efficiency of the proposed controller.

This paper addresses the problem of controlling multiple unmanned aerial vehicles (UAVs) cooperating in a formation to carry out a complex task such as surface inspection. We first use the virtual leader-follower model to determine the topology and trajectory of the formation. A double-loop control system combining backstepping and sliding mode control techniques is then designed for the UAVs to track the trajectory. A radial basis function neural network capable of estimating external disturbances is developed to enhance the robustness of the controller. The stability of the controller is proven by using the Lyapunov theorem. A number of comparisons and software-in-the-loop tests have been conducted to evaluate the performance of the proposed controller. The results show that our controller not only outperforms other state-of-the-art controllers but is also sufficient for complex tasks of UAVs such as collecting surface data for inspection. The source code of our controller can be found at https://github.com/duynamrcv/rbf_bsmc.

The reconfigurable mechanisms can satisfy the requirements of changing environments, working conditions, and tasks on the function and performance of the mechanism and can be applied to machine tool manufacturing, space detection, etc. Inspired by the single-vertex fivefold origami pattern, a new reconfigurable parallel mechanism is proposed in this paper, which has special singular positions and stable motion due to replicating the stabilizing kinematic properties of origami. Through analyzing the topologic change of the folding process of the pattern and treating it as a reconfigurable joint, a new reconfigurable parallel mechanism with 3, 4, 5, or 6 degrees of freedom is obtained. Then, the kinematics solution, workspace, and singularity of the mechanism are calculated. The results indicate that the singular configuration of the origami-derived reconfigurable parallel mechanism is mainly located in a special plane, and the scope of the workspace is still large after the configuration change. The mechanism has the potential to adapt to multiple tasks and working conditions through the conversion among different configurations by folding reconfigurable joints on the branch chain.

Autonomous navigation has been a long-standing research topic, and researchers have worked on many challenging problems in indoor and outdoor environments. One application area of navigation solutions is material handling in industrial environments. With Industry 4.0, the simple problem in traditional factories has evolved into the use of autonomous mobile robots within flexible production islands in a self-decision-making structure. Two main stages of such a navigation system are safe transportation of the vehicle from one point to another and reaching destinations at industrial standards. The main concern in the former is roughly determining the vehicle’s pose to follow the route, while the latter aims to reach the target with high accuracy and precision. Often, it may not be possible or require extra effort to satisfy requirements with a single localization method. Therefore, a multi-stage localization approach is proposed in this study. Particle filter-based large-scale localization approaches are utilized during the vehicle’s movement from one point to another, while scan-matching-based methods are used in the docking stage. The localization system enables the appropriate approach based on the vehicle’s status and task through a decision-making mechanism. The decision-making mechanism uses a similarity metric obtained through the correntropy criterion to decide when and how to switch from large-scale localization to precise localization. The feasibility and performance of the developed method are corroborated through field tests. These evaluations demonstrate that the proposed method accomplishes tasks with sub-centimeter and sub-degree accuracy and precision without affecting the operation of the navigation algorithms in real time.

This paper mainly studies an autonomous path-planning and real-time path-tracking optimization method for snake robot. Snake robots can perform search and rescue, exploration, and other tasks in a variety of complex environments. Robots with visual sensors such as LiDAR can avoid obstacles in the environment through autonomous navigation to reach the target point. However, in an unstructured environment, the navigation of snake robot is easily affected by the external environment, causing the robot to deviate from the planned path. In order to solve the problem that snake robots are easily affected by environmental factors in unstructured environments, resulting in poor path-following ability, this paper uses the Los algorithm combined with steering control to plan the robot in real time and control the robot’s steering parameters in real time, ensuring that the robot can stably follow the planned path.

This paper deals with generally routed, pre-bent cable-driven continuum robots (CCR). A CCR consists of a flexible backbone to which multiple disks are attached. Cables are passed through holes in the disk, and when pulled, the flexible backbone and the CCR can attain different shapes based on their routing and backbone configuration. An optimization-based approach, using minimization of strain energy, is shown to give good results for the pose and motion of the CCR and to determine contact with external objects. The pose, motion, and the contact obtained from the model are shown to match very well with experimental results obtained from a 3D-printed CCR. An algorithm is proposed to generate the pre-bent backbone for a CCR which on actuation can attain the desired shape. Using the algorithm, three 3D-printed CCRs with pre-bent backbones are fabricated and these are used to demonstrate a compliant gripper that can grip a spherical object similar to that done by tentacles, and another three-fingered gripper with straight backbone CCRs is used to orient a square object gripped at the end.

This paper proposes a virtual reality-based dual-mode teleoperation architecture to assist human operators in remotely operating robotic manipulation systems in a safe and flexible way. The architecture, implemented via a finite state machine, enables the operator to switch between two operational modes: the Approach mode, where the operator indirectly controls the robotic system by specifying its target configuration via the immersive virtual reality (VR) interface, and the Telemanip mode, where the operator directly controls the robot end-effector motion via input devices. The two independent control modes have been tested along the task of reaching a glass on a table by a sample population of 18 participants. Two working groups have been considered to distinguish users with previous experience with VR technologies from the novices. The results of the user study presented in this work show the potential of the proposed architecture in terms of usability, both physical and mental workload, and user satisfaction. Finally, a statistical analysis showed no significant differences along these three metrics between the two considered groups demonstrating ease of use of the proposed architecture by both people with and with no previous experience in VR.

In an effort to alleviate the issue of knee joint fatigue and injury during lower limb ambulation, a novel passive assisted exoskeleton robot with human–machine interaction is investigated to assist the movement of the human knee joint. The design of the exoskeleton configuration takes into consideration the physiological structure and gait function of the knee joint, ensuring that it satisfies the requirements for motion, force, and gait function of the knee joint. To explore the interaction between the wearer and the exoskeleton, a human–machine kinematic model after wearing exoskeleton is established, which is instrumental in analyzing the integration motion of the wearer and exoskeleton. In addition, the dynamic and static models of the knee joint after wearing the exoskeleton are established, utilizing the Newton–Euler method and force polygon method, respectively, to evaluate the effectiveness of the exoskeleton. Moreover, the size parameters and spring stiffness of the exoskeleton are optimized, using both human body kinematic model and mechanic model. Furthermore, the effectiveness of the exoskeleton in proving assistance is evaluated through human body simulation, using OpenSim. The results indicate that the exoskeleton significantly reduces the knee joint torque by 48.42%.

In this paper, we propose a novel vision-based adaptive leakage-type (LT) sliding mode admittance control for actuator-constrained collaborative robots to realize the synchronous control of the precise path following and compliant interaction force. Firstly, we develop a vision-admittance-based model to couple the visual feedback and force sensing in the image feature space so that a reference image feature trajectory can be obtained concerning the contact force command and predefined trajectory. Secondly, considering the system uncertainty, external disturbance, and torque constraints of collaborative robots in reality, we propose an adaptive sliding mode controller in the image feature space to perform precise trajectory tracking. This controller employs a leakage-type (LT) adaptive control law to reduce the side effects of system uncertainties without knowing the upper bound of system uncertainties. Moreover, an auxiliary dynamic is considered in this controller to overcome the joint torque constraints. Finally, we prove the convergence of the tracking error with the Lyapunov stability analysis and operate various semi-physical simulations compared to the conventional adaptive sliding mode and parallel vision/force controller to demonstrate the efficacy of the proposed controller. The simulation results show that compared with the controller mentioned above, the path following accuracy and interaction force control precision of the proposed controller increased by 50% and achieved faster convergence.

This paper presents a concurrent optimization approach for the design and motion of a quadruped in order to achieve energy-efficient cyclic behaviors. Computational techniques are applied to improve the development of a novel quadruped prototype. The scale of the robot and its actuators are optimized for energy efficiency considering the complete actuator model including friction, torque, and bandwidth limitations. This method and the optimal bounding trajectories are tested on the first (non-optimized) prototype design iteration showing that our formulation produces a trajectory that (i) can be easily replayed on the real robot and (ii) reduces the power consumption w.r.t. hand-tuned motion heuristics. Power consumption is then optimized for several periodic tasks with co-design. Our results include, but are not limited to, a bounding and backflip task. It appears that, for jumping forward, robots with longer thighs perform better, while, for backflips, longer shanks are better suited. To explore the tradeoff between these different designs, a Pareto set is constructed to guide the next iteration of the prototype. On this set, we find a new design, which will be produced in future work, showing an improvement of at least 52% for each separate task.

In this paper, a hybrid approach organized in four phases is proposed to solve the multi-objective trajectory planning problem for industrial robots. In the first phase, a transcription of the original problem into a standard multi-objective parametric optimization problem is achieved by adopting an adequate parametrization scheme for the continuous robot configuration variables. Then, in the second phase, a global search is performed using a population-based search metaheuristic in order to build a first approximation of the Pareto front (PF). In the third phase, a local search is applied in the neighborhood of each solution of the PF approximation using a deterministic algorithm in order to generate new solutions. Finally, in the fourth phase, results of the global and local searches are gathered and postprocessed using a multi-objective direct search method to enhance the quality of compromise solutions and to converge toward the true optimal PF. By combining different optimization techniques, we intend not only to improve the overall search mechanism of the optimization strategy but also the resulting hybrid algorithm should keep the robustness of the population-based algorithm while enjoying the theoretical properties of convergence of the deterministic component. Also, the proposed approach is modular and flexible, and it can be implemented in different ways according to the applied techniques in the different phases. In this paper, we illustrate the efficiency of the hybrid framework by considering different techniques available in various numerical optimization libraries which are combined judiciously and tested on various case studies.

This article presents the development of a robot capable of modifying its size through a wheel reconfiguration strategy. The reconfigurable wheel design is based on a four-bar retractable mechanism that achieves variation of the effective radius of the wheel. A reconfiguration index is introduced based on the number of retractable mechanisms that predicts the radius of configuration according to the number of mechanisms implemented in the wheel. The kinematics of the retractable mechanism is studied to determine the theoretical reconfiguration radius during the transformation process, it is also evaluated numerically with the help of the GeoGebra software, and it is validated experimentally by image analysis using the Tracker software. The transformation process of the robot is investigated through an analysis of forces that consider the wheel in contact with the obstacle, the calculation of the wheel torque and the height of the obstacle to be overcome are presented. On the other hand, the experimental validation of the robot reconfiguration process is presented through the percentage of success shown by the robot to overcome obstacles of 50, 75, 100 and 125 mm. In addition, measurements of energy consumption during the transformation process are reported. Reconfigurable wheels, capable of adapting their size, offer innovative solutions to various challenges across different applications such as robotic exploration and search and rescue missions to industrial settings. Some key issues that these wheels can address include terrain adaptability enhancing a robot’s mobility over uneven surfaces, or obstacles; enhanced robotic design; cost-effective design; space efficiency; and versatility in applications.