Human gait trajectory prediction is a long-standing research topic in human–machine interaction. However, there are two shortcomings in the current gait trajectory prediction technology. The first shortcoming is that the neural network model of gait prediction only predicts dozens of future time frames of gait trajectory. The second shortcoming is that the gait prediction neural network model is uninterpretable. We propose the Interpretable-Concatenation former (IC-former) model, which can predict long-term gait trajectories and explain the prediction results by quantifying the importance of data at different positions in the input sequence. Experiments prove that the IC-former model we proposed not only makes a breakthrough in prediction accuracy but also successfully explains the data basis of the prediction.

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 explores and experimentally compares the effectiveness of robot-stopping approaches based on the speed and separation monitoring for improving fluency in collaborative robotics. In the compared approaches, a supervisory controller checks the distance between the bounding volumes enclosing human operator and robot and prevents potential collisions by determining the robot’s stop time and triggering a stop trajectory if necessary. The methods are tested on a Franka Emika robot with 7 degrees of freedom, involving 27 volunteer participants, who are asked to walk along assigned paths to cyclically intrude the robot workspace, while the manipulator is working. The experimental results show that scaling online the dynamic safety zones is beneficial for improving fluency of human-robot collaboration, showing significant statistical differences with respect to alternative approaches.

The ability of quadruped robots to overcome obstacles is a critical factor that limits their practical application. Here, a design concept and a control algorithm are presented that aim at enhancing the explosive force of quadruped robots during jumping by utilizing elastic energy storage components. The hind legs of the quadruped robot are designed as energy storage units. Tension springs are utilized as components for storing energy and are installed in a parallel structure on the hind leg. Energy is stored during the compression process of the robot’s torso and released during the jumping phase. The optimal foot force is calculated using a single rigid body model. The mapping relationship between the force applied to the foot and the resulting joint torque is established by developing a dynamic model of the hind legs. Simulation experiments were conducted using the Webots physics engine to compare the impact of varying spring stiffness on joint torque during the jumping process. This study determined the optimal spring stiffness under specific conditions. The hind legs’ torque saving ratio reaches 19%, and the energy-saving ratio reaches 13%, which validates the effectiveness and feasibility of integrating elastic energy storage components.

When generating simultaneous joint movements of a humanoid with multiple degrees of freedom to replicate human-like movements, the approach of joint synergy can facilitate the generation of whole-body robotic movement with a reduced number of control inputs. However, the trade-off of minimizing control inputs and keeping characteristics of movements makes it difficult to improve movement performance in a simple control manner. In this paper, we introduce an approach by connecting and constraining these joints. It is inspired by the fascia network of the human body, which constrains the whole-body movements of a human. Compared to when only joint synergy is used, the effectiveness of the proposed method is verified by calculating the errors of joint positions of generated movements and human movements. The paper provides a detailed exploration of the proposed method, presenting simulation-experimental results that affirm its effectiveness in generated movements that closely resemble human movements. Furthermore, we provide one possible method on how these concepts can be implemented in actual robotic hardware, offering a pathway to improve movement control in humanoid robots within their mechanical limitations.

The intricate water-land intermingled nature of wild environments necessitates robots to exhibit multimodal cross-domain mobility capabilities. This paper introduces a novel wheel-spoke-paddle hybrid amphibious robot (WSP-bot) that can operate on flat and rough terrains, water surfaces, and water-land transitional zones. The proposed robot relies on a propulsion mechanism called transformable wheel-spoke-paddle (WSP), which combines the stability of wheeled robots with the obstacle-climbing capability of legged robots, while also providing additional aquatic mobility. The utilization of a crank-slider-based transformation mechanism enables seamless switching between multiple motion modes. An analysis of mode transition and ground motion in spoke mode was conducted, along with an investigation of its obstacle-crossing capability. Simulations were performed for mode transition, ground locomotion, and obstacle-crossing, as well as propulsion of a single WSP module on water. Based on the above work, a prototype robot was manufactured. Prototype tests, including mode transition and mobility tests on land and water surfaces under multimodal states, confirmed the effectiveness of the proposed WSP-bot.

An online pattern recognition method of lower limb movements is proposed based on the personalized surface electromyography (sEMG) signals, and the corresponding experimental researches are performed in the rehabilitation training. Further, a wireless wearable acquisition instrument is used. Based on this instrument, a host computer for the personal online recognition and real-time control of rehabilitation training is developed. Three time-domain features and two features in the nonlinear dynamics are selected as the joint set of the characteristic values for the sEMG signals. Then a particle swarm optimization (PSO) algorithm is used to optimize the feature channels, and a k-nearest neighbor (KNN) algorithm and the extreme learning machine (ELM) algorithm are combined to classify and recognize individual sample data. Based on the multi-pose lower limb rehabilitation robot, the real-time motion recognition and the corresponding rehabilitation training are carried out by using the online personalized classifier. The experimental results of eight subjects indicate that it takes only 6 min to build an online personalized classifier for the four types of the lower limb movements. The recognition between switches of different rehabilitation training movements is timely and accurate, with an average recognition accuracy of more than 95%. These results demonstrate that this system has a strong practicability.

Heavy-duty hexapod robots are well-suited for physical transportation, disaster relief, and resource exploration. The immense locomotion capabilities conferred by the six appendages of these systems enable traversal over unstructured and challenging terrain. However, tipping can be a serious concern when moving with a tripod gait in these challenging environments, which may cause irreversible consequences such as compromised movement control and potential damage. In this paper, we focus on heavy-duty hexapod robot sideline tipping judgment and recovery during tripod gait motion, and a novel sideline tipping judgment and recovery method is proposed by adjusting an optimal swinging leg to the stance state. Considering the locomotion environments, motion mode, and tipping analysis, the robot’s stability margin is quantified, and the tipping event is evaluated by the Force Angle Stability Measure (FASM). The recovery method is initiated upon detecting that the robot is tipping, which involves the selection of an adjustment leg and the determination of an optimal foothold. Since the FASM is based on the foot force and robot center of gravity (CoG), the stability margin quantification expression is reformulated to the constraint form of quadratic programming (QP). Furthermore, a foot force distribution method, integrating stability margin considerations into the QP model, has been devised to ensure post-adjustment stability of the landing leg. Experiments on tipping judgment and recovery demonstrate the effectiveness of the proposed approaches on tipping judgment and recovery.

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.

The integration of camera and LiDAR technologies has the potential to significantly enhance construction robots’ perception capabilities by providing complementary construction information. Structured light cameras (SLCs) are a desirable alternative as they provide comprehensive information on construction defects. However, fusing these two types of information depends largely on the sensors’ relative positions, which can only be established through extrinsic calibration. This paper introduces a novel calibration algorithm considering a customized board for SLCs and repetitive LiDARs, which are designed to facilitate the automation of construction robots. The calibration board is equipped with four symmetrically distributed hemispheres, whose centers are obtained by fitting the spheres and adoption with the geometric constraints. Subsequently, the spherical centers serve as reference features to estimate the relationship between the sensors. These distinctive features enable our proposed method to only require one calibration board pose and minimize human intervention. We conducted both simulation and real-world experiments to assess the performance of our algorithm. And the results demonstrate that our method exhibits enhanced accuracy and robustness.

This paper proposes a kinematic calibration method of a novel 5-degree-of-freedom double-driven parallel mechanism with the sub-closed loop on limbs. At first, considering the introduction of a sub-closed loop significantly increased the complexity and difficulty of kinematic error modeling, an equivalent transformation method is proposed for the limb with a sub-closed loop. Then kinematic error model of the parallel mechanism is established based on the closed-loop vector method and parasitic motion analysis, which is verified by virtual prototype technology. Because the full kinematic error model is generally redundant, error parameter identifiability analysis is carried out by QR decomposition of the identification Jacobian matrix, and the redundant parameters are removed. Additionally, the Sequence Forward Floating Search algorithm is utilized to optimize measurement configurations to reduce the influence of measurement noise. Finally, with a laser tracker as the measuring device, numerical simulations and experiments are implemented to verify the proposed kinematic calibration method. The experiment results show that average position and orientation errors are reduced from 2.778 mm and 1.115° to 0.263 mm and 0.176°, respectively, within the prescribed workspace.

Parallel manipulators with flexible morphing platform (FMP) provide potential solution in various application fields, such as shape-morphing underwater robot, deformable wings, and human–machine interfaces. However, there is still lack of effective approach for the design and analysis of such novel type of parallel manipulator. In this article, a 9-UPS redundant actuation parallel manipulator with flexible morphing moving platform is designed as a representative of this kind of manipulator. Correspondingly, a deformation estimation and shape control approach for the FMP is presented. The proposed deformation estimation approach is designed based on the bending energy, which can achieve high calculation efficiency and avoid complex mechanical definition and calculation. And the proposed shape control approach is realized by utilizing a nonrigid ICP match algorithm, which can continuously deform the morphing platform to an arbitrary target surface. A prototype of the 9-UPS parallel manipulator is fabricated and analyzed as verification. The experiment results show that the proposed approach offers a promising avenue for the deformation estimation and shape control of the morphing platform.

In confined multi-obstacle environments, generating feasible paths for continuum robots is challenging due to the need to avoid obstacles while considering the kinematic limitations of the robot. This paper deals with the path-planning algorithm for continuum robots in confined multi-obstacle environments to prevent their over-deformation. By modifying the tree expansion process of the Rapidly-exploring Random Tree Star (RRT*) algorithm, a path-planning algorithm called the continuum-RRT* algorithm herein is proposed to achieve fewer iterations and faster convergence as well as generating desired paths that adhere to the kinematic limitations of the continuum robots. Then path planning and path tracking are implemented on a tendon-driven four-section continuum robot to validate the effectiveness of the path-planning algorithm. The path-planning results show that the path generated by the algorithm indeed has fewer transitions, and the path generated by the algorithm is closer to the optimal path that satisfies the kinematic limitations of the continuum robot. Furthermore, path-tracking experiments validate the successful navigation of the continuum robot along the algorithm-generated path, exhibiting an error range of 2.51%–3.91%. This attests to the effectiveness of the proposed algorithm in meeting the navigation requirements of continuum robots.

Currently, workers in sand casting face harsh environments and the operation safety is poor. Existing pouring robots have insufficient stability and load-bearing capacity and cannot perform intelligent pouring according to the demand of pouring process. In this paper, a hybrid pouring robot is proposed to solve these limitations, and a vision-based hardware-in-the-loop (HIL) control technology is designed to achieve the real-time control problems of simulated pouring and pouring process. Firstly, based on the pouring mechanism and the motion demand of ladle, a hybrid pouring robot with a 2UPR-2RPU parallel mechanism as the main body is designed. And the equivalent hybrid kinematic model was established by using Eulerian method and differential motion. Subsequently, a motion control strategy based on HIL simulation technique was designed and presented. The working space of the robot was obtained through simulation experiments to meet the usage requirements. And the stability of the robot was tested through the key motion parameters of the robot joints. Based on the analysis of pouring quality and trajectory, optimal dynamic parameters for the experimental prototype are obtained through water simulation experiments, the pouring liquid height area is 35–40 cm, the average flow rate of pouring liquid is 112 cm3/s, and the ladle tilting speed is 0.0182 rad/s. Experimental results validate the reasonableness of the designed pouring robot structure. Its control system realizes the coordinated movement of each branch chain to complete the pouring tasks with different variable parameters. Consequently, the designed pouring robot will significantly enhance the automation level of the casting industry.

This article proposes a control method for underactuated cartpole systems using semi-implicit cascaded proportional-derivative (PD) controller. The proposed controller is composed of two conventional PD controllers, which stabilizes the pole and the cart second-order dynamics respectively. The first PD controller is realized by transforming the pole dynamics into a virtual PD controller, with the coupling term exploited as the internal tracking target for the cart dynamics. Then, the second PD controller manipulates the cart dynamics to track that internal target. The solution to the internal tracking target relies on an equation set and features a semi-implicit process, which exploits the internal dynamics of the system. Besides, the design of second PD controller relies on the parameters of the first PD controller in a cascaded manner. A stability analysis approach based on Jacobian matrix is proposed and implemented for this fourth-order system. The proposed method is simple in design and intuitive to comprehend. The simulation results illustrate the superiority of proposed method compared with conventional double-loop PD controller in terms of convergence, with the theoretical conclusion of at least locally asymptotic stability.

The widely used model predictive control of discrete-time control barrier functions (MPC-CBF) has difficulties in obstacle avoidance for unmanned ground vehicles (UGVs) in complex terrain. To address this problem, we propose adaptive dynamic control barrier functions (AD-CBF). AD-CBF is able to adaptively select an extended class of functions of CBF to optimize the feasibility and flexibility of obstacle avoidance behaviors based on the relative positions of the UGV and the obstacle, which in turn improves the obstacle avoidance speed and safety of the MPC algorithm when integrated with MPC. The algorithmic constraints of the CBF employ hierarchical density-based spatial clustering of applications with noise (HDBSCAN) for parameterization of dynamic obstacle information and unscaled Kalman filter (UKF) for trajectory prediction. Through simulations and practical experiments, we demonstrate the effectiveness of the AD-CBF-MPC algorithm in planning optimal obstacle avoidance paths in dynamic environments, overcoming the limitations of the point-by-point feasibility of MPC-CBF.

Most of the currently available three-degree-of-freedom manipulators are light load and cannot achieve full continuous rotation; given this, we designed a heavy-load manipulator that achieves unrestricted and continuous rotation. Due to manufacturing and assembly errors, parameter deviations between the real manipulator and its underlying theoretical model were unavoidable. Because of the lack of high-precision, high-frequency, and real-time closed-loop detection methods, we proposed a type of kinematics calibration of parameterized ant colony optimization and feedforward control methods. This was done to achieve high-precision motion control. First, an error model combining structural parameters and joint output angles was established, and the global sensitivity of each error source was analyzed to distinguish both primary and secondary sources. Based on the measured data of a laser tracker, the ant colony optimization was then used to identify six error sources. This resulted in both link length and joint driving errors of the designed manipulator. As it is a type of systematic error, the rounding error of the theoretical trajectory was carefully analyzed, and feedforward control methods with different coefficients were designed to further improve positioning accuracy based on the kinematic calibration. Experimental results showed that the proposed kinematic calibration and feedforward control methods achieved relatively precise motion control for the designed manipulator.

Robots with multi-sensors always have a problem of weak pairing among different modals of the collected information produced by multi-sensors, which leads to a bad perception performance during robot interaction. To solve this problem, this paper proposes a Force Vision Sight (FVSight) sensor, which utilizes a distributed flexible tactile sensing array integrated with a vision unit. This innovative approach aims to enhance the overall perceptual capabilities for object recognition. The core idea is using one perceptual layer to trigger both tactile images and force-tactile arrays. It allows the two heterogeneous tactile modal information to be consistent in the temporal and spatial dimensions, thus solving the problem of weak pairing between visual and tactile data. Two experiments are specially designed, namely object classification and slip detection. A dataset containing 27 objects with deep presses and shallow presses is collected for classification, and then 20 slip experiments on three objects are conducted. The determination of slip and stationary state is accurately obtained by covariance operation on the tactile data. The experimental results show the reliability of generated multimodal data and the effectiveness of our proposed FVSight sensor.

Unmanned aerial vehicles (UAVs) possess fast-moving abilities and have been used in various tasks in the past decades. However, their performances are still restricted by insufficient endurance and confined environments. Intuitively, combining other locomotion modes with UAVs, such as diving and driving, would be an appropriate idea to improve the robot’s adaptability and solve the endurance problem. Recently, the terrestrial/aerial hybrid robots have drawn the researchers’ eyes for their outstanding performances, which can deploy flight mode to traverse insurmountable terrains and ground mode to increase endurance and realize detailed searches. Therefore, this paper developed the autonomous quadrotor tilting hybrid robot (AQT-HR) to achieve terrestrial/aerial dual-modal mobility and verified that the robot delivers high energy efficiency. The AQT-HR can achieve flying and driving through a quadrotor tilting mechanism, which can alter one single driving force into different directions. Furthermore, the dynamic models of the hybrid robot’s aerial and ground locomotion are derived and introduced into the model-feedforward PID control algorithm for improving the robot’s flying stability. Finally, we conducted some mobility tests and experiments about traversing obstacles to demonstrate that the proposed hybrid robot can realize autonomous mode switching and perform a low energy consumption in ground movement mode.

The language-guided visual robotic grasping task focuses on enabling robots to grasp objects based on human language instructions. However, real-world human-robot collaboration tasks often involve situations with ambiguous language instructions and complex scenarios. These challenges arise in the understanding of linguistic queries, discrimination of key concepts in visual and language information, and generation of executable grasping configurations for the robot’s end-effector. To overcome these challenges, we propose a novel multi-modal transformer-based framework in this study, which assists robots in localizing spatial interactions of objects using text queries and visual sensing. This framework facilitates object grasping in accordance with human instructions. Our developed framework consists of two main components. First, a visual-linguistic transformer encoder is employed to model multi-modal interactions for objects referred to in the text. Second, the framework performs joint spatial localization and grasping. Extensive ablation studies have been conducted on multiple datasets to evaluate the advantages of each component in our model. Additionally, physical experiments have been performed with natural language-driven human-robot interactions on a physical robot to validate the practicality of our approach.