This paper presents a study on the effect of design parameters of an underactuated hand on its grasp performance. Three kinds of grasp performance characteristics are considered: grasp range, grasp strength, and immobilizing grasp range. Grasp strength is defined as the stiffness of the grasp. Immobilizing grasps are those in which the object cannot be moved for a force up to a certain threshold. In general, underactuated hands cannot produce immobilizing grasps. However, we show that immobilizing grasps can be created by including joint limits in the hand design. We consider the effect of two design parameters on the grasp performance: torque ratio and finger-base distance. Results show that an optimal finger-base distance and torque ratio exists that maximizes the grasp range and grasp strength. Also, the immobilizing grasp range is increased by decreasing the finger-base distance and increasing the torque ratio and joint limits.

This paper presents a new feature based approach for place recognition using 2D LiDAR (Light Detection And Ranging) data. The main contribution lies in the mapping process. It includes a keypoint selection strategy to model places with persistent keypoints from concatenated LiDAR scans. Our objective is to achieve map data reduction while maintaining good place recognition performace. LiDAR scans are concatenated with a registration algorithm and keypoints are extracted from each scan. Based on a regular grid, our approach measures the occurrence of similar keypoints in each region of interest defined by a grid cell. Only keypoints with occurrences beyond a threshold, qualified as stable keypoints, are kept in the place model called submap. The environment is therefore modeled as a collection of submaps, which constitutes the global map. Place recognition consists in submap matching followed by a two steps geometric verification. In the first stage, optimal parameters are set using corrected data. Mapping parameters satisfy six criteria among which is the spatial distribution distance, which represents another contribution of our work. It gives a measure on how well keypoints are distributed in space. Place recognition optimal parameters are set through global localization. In the second stage, the approach is evaluated using raw data in the contexts of global localization, and loop closure detection in a SLAM framework. The results obtained using popular real data sets show that our approach achieves significant map data reduction (up to $92\%$ ) while maintaining good place recognition performance, comparable to state-of-the-art methods.

In this paper, we propose a set of robust training methods for deep reinforcement learning to transfer learning acquired in one control task to a set of previously unseen control tasks. We improve generalization in commonly used transfer learning benchmarks by a novel sample elimination technique, early stopping, and maximum entropy adversarial reinforcement learning. To generate robust policies, we use sample elimination during training via a method we call strict clipping. We apply early stopping, a method previously used in supervised learning, to deep reinforcement learning. Subsequently, we introduce maximum entropy adversarial reinforcement learning to increase the domain randomization during training for a better target task performance. Finally, we evaluate the robustness of these methods compared to previous work on simulated robots in target environments where the gravity, the morphology of the robot, and the tangential friction coefficient of the environment are altered.

To set the sleeping mode for the Yutu-2 rover, a visual pose prediction algorithm including terrain reconstruction and pose estimation was first studied. The terrain reconstruction precision is affected by using only the stereo navigation camera (Navcam) images and the rotation angles of the mast. However, the hazard camera (Hazcam) pose is fixed, and an image network was constructed by linking all of the Navcam and Hazcam stereoimages. Then, the Navcam pose was refined based on a multiview block bundle adjustment. The experimental results show that the mean absolute errors of the check points in the proposed algorithm were 10.4 mm over the range of $\boldsymbol{L}$ from 2.0 to 6.1 m, and the proposed algorithm achieved good prediction results for the rover pose (the average differences of the values of the pitch angle and the roll angle were −0.19 degrees and 0.29 degrees, respectively). Under the support of the proposed algorithm, engineers have completed the remote setting of the sleeping mode for Yutu-2 successfully in the Chang’e-4 mission operations.

In this paper, control of a suspended cable-driven parallel robot has been experimentally investigated based on the dynamic model of the robot for object tracking purpose. In order to improve the tracking ability of the robot, three control approaches, namely kinematic PID, dynamic PD, and a kinematic sliding mode control (SMC), have been implemented, both on the Simscape and on the robot constructed at the Human and Robot Interaction Laboratory. Neural network controller and dynamic SMC have been implemented on the Simscape model. Afterward, the effectiveness of each approach has been investigated by employing the root mean square error (RMSE) index. Simulation and experimental results reveal the ability of each controller for precise and smooth control. For precise and real-time object tracking, YOLOv5-s and YOLOv4-tiny model are trained. By comparing the obtained index values, the kinematic PID demonstrates the best performance with the maximum RMSE value of 0.018 compared to other methods.

Robotic systems are usually controlled to repetitively perform specific actions for manufacturing tasks. The traditional control methods are domain-dependent and model-dependent with cost of much human efforts. They cannot meet the new requirements of generality and flexibility in many areas such as intelligent manufacturing and customized production. This paper develops a general model-free approach to enable robots to perform multi-step object sorting tasks through deep reinforcement learning. Taking projected heightmap images from different time steps as input without extra high-level image analysis and understanding, critic models are designed to produce a pixel-wise Q value map for each type of action. It is a new trial to apply pixel-wise Q value-based critic networks to solve multi-step sorting tasks that involve many types of actions and complex action constraints. The experimental validations on simulated and realistic object sorting tasks demonstrate the effectiveness of the proposed approach. Qualitative results (videos), code for simulated and realistic experiments, and pre-trained models are available at https://github.com/JiatongBao/DRLSorting

This paper proposes a robot model with multiple spines and investigates its effects on the bounding gait of quadruped robot. Contrastive tests, carried out by adding an active driving joint (ADJ) and changing the number of passive driving joints, were divided between simulation and physical. The results reveal that the spinal structure (one ADJ and several passive driving joints) is closely related to the performance parameters. Increasing the number of passive driving joints can improve the velocity and energy efficiency. This structure may have significant guidance for designing a quadruped robot.

Soft crawling robots have been significantly studied in recent decades. However, moving in amphibious environment, high payload capability, and passing through complex ground have always been challenges for soft crawling robots. For these problems, this article presents an amphibious soft-rigid wheeled crawling robot (SRWCR) consists of a soft-rigid body actuated by two soft pneumatic actuators (SPAs), four wheels, and four annular soft bladders (ASBs) as brakes. By programming the actuation sequences of the two SPAs and four ASBs, SRWCR can achieve two basic modes of locomotion: linear motion and turning. Based on the energy conservation law, we have developed analytical models to interpret the static actuation performance of SPA, including linear and bending deformations. Furthermore, with the help of fast response and waterproof of SPA and ASB, SRWCR can achieve a linear speed of 14.97 mm/s, a turning speed of 5.63°/s, and an underwater locomotion speed of 13 mm/s, which demonstrates the excellent locomotion performance of SRWCR in amphibious environment. In addition, SRWCR can also achieve multiple impressive functions, including carrying a payload of 2 kg at the moving speed of 11.18 mm/s, passing through various complex ground such as the grass ground and sand ground, and so on, obstacle navigation in confined space. Compared with the existing soft crawling robots, with the help of the soft-rigid body and wheeled structure, SRWCR has the best payload and passing capability, which indicates the potential advantage of SRWCR in the design of functional robots.

Robotics with artificial intelligence techniques have been the center of attraction among researchers as it is well equipped in the area of human intervention. Here, the krill herd (KH) optimization algorithm is modified and hybridized with a fuzzy logic controller to frame an intelligent controller for optimal trajectory planning and control of mobile robots in obscure environments. The controller is demonstrated for single and multiple robot’s trajectory planning. A Petri-net controller has also been added to avoid conflict situations in multi-robot navigation. MATLAB and V-REP software are used to simulate the work, backed with real-time experiments under laboratory conditions. The robots efficiently achieved the goals by tracing an optimal path without any collision. Trajectory length and time spent during navigation are recorded, and a good agreement between the results is observed. The proposed technique is compared against existing research techniques, and an improvement of 14.26% is noted in terms of path length.

The rotational joint plays a vital role in the industrial and civil areas, which is typically utilized to achieve the relative rotation between the adjacent parts. Generally, structuring a conventional rotational joint involves the bulk actuators (e.g., motor and hydraulic cylinders) and complex structures, bringing difficulty for miniaturing the dimension. In this paper, a class of novel rotational mechanisms, which were constructed by the combination of compliant mechanisms (e.g., cartwheel pivot and multileaf pivot) and intelligent actuator (e.g., shape memory alloy (SMA) wire and spring), was proposed to reduce the complexity of the conventional rotational joints. As the case study, a novel SMA wire-driven flexural rotational mechanism (SDFRM), which is constructed by the cartwheel pivot and SMA wire, was developed to demonstrate the feasibility of combining the compliant mechanism and smart actuator. After establishing the static model of the cartwheel pivot and the thermal effect model of the SMA wire, the overall model of SDFRM was developed for the comprehensive performance analysis and the control system design. After that, the model validation and experiments were performed with the proposed prototype and control system. It can be seen from the experimental results that the proposed model can be validated within the error of 3.8%. In addition, the performance study on SDFRM indicates that the prototyped SDFRM system can track the given trajectories within the error of 0.2 mm in the workspace. As a result, the proposed concept was demonstrated as an effective way to reduce the dimension and weight of the conventional rotational joint.

In the paper, a novel modular soft robot that can crawl and turn is presented. The modular soft robot is composed of multiple drive modules connected in series, including one head module, one tail module and three body modules. Each module is actuated by the air chamber. Due to the nonlinear performance of the air chamber, the strain energy function of the air chamber is established. The relationship between the displacement of the air chamber expansion wall and the inflation pressure is obtained, and the manufacturing parameters of the air chamber are determined. By dividing the body of the robot into a series of continuous flexible models, the driving force and the friction force of the robot in locomotion are analyzed. An inflation and deflation control method is presented to complete the locomotion. According to the experiment, the crawling speed of the robot can reach 15.53 mm/s (0.03 body length per second). The turning speed of the robot can reach 1.273 °/s. The robot can crawl and turn on the rough blanket surface effectively. The robot can explore and move in a complex and changeable environment.

Aiming at the problem that adaptive Monte Carlo localization (AMCL) algorithm is difficult to localize in large scenes and similar environments. This paper uses a semantic information-assisted approach to improve the AMCL algorithm. This method realizes the robust localization of the robot in the large scenes and similar environments. Firstly, the 2D grid map created by simultaneous localization and mapping using lidar can obtain highly accurate indoor environmental contour information. Secondly, the semantic object capture is achieved by using a depth camera combined with an instance segmentation algorithm. Then, the semantic grid map is created by mapping the semantic point cloud through the back-projection process of the pinhole camera. Finally, semantic grid maps are used as a priori information to assist in localization, which will be used to improve the initial particle swarm distribution in the global localization of the AMCL algorithm and thus will solve the robot localization problem in this environment. The experimental evidence shows that the semantic grid map solves the environmental information degradation problem caused by 2D lidar as well as improves the robot’s perception of the environment. In addition, this paper improves the localization robustness of the AMCL algorithm in large scenes and similar environments, resulting in an average localization success rate of about 90% or even higher, and further reduces the number of iterations. The global localization problem of robots in large scenes and similar environments is effectively solved.

The principal objective of the paper is to show the importance of the Hamiltonian in control theory. Instead of using the Lagrangian formulation of electromechanical or robotic systems, our work is focused on robot dynamics by its Hamiltonian. Using the iterative Newton–Euler, we generate the local Hamiltonians and the derivative of the moments at each joint of the robot manipulator. Thus, we can apply decentralized controllers at each joint. We compare and discuss the efficiency of the controllers. We show that the performance of the sliding modes controller is more robust than that of the PD or Bang–Bang controllers.

For an underactuated anthropomorphic hand, apart from replicating the geometry, biomimicry of human counterpart requires design based on functional biomimesis. Optimization based on grasping capabilities to generate a stable grasp for underactuated anthropomorphic hands has been a recent focus. In this article, we optimize the actuation parameters of the underactuated mechanism based on grasp quality measures. Optimization of the hand design parameters like pulley radii, spring stiffnesses, and pre-load angles was undertaken based upon the grasp robustness metric. A quasi-static model based on a soft synergistic compliant grasp of the underactuated hand is formalized. Numerical simulations based on evolutionary strategies are applied to the grasp model to optimize the underactuated parameters. Finally, validation of the results based on the grasp wrench space is presented. The results show that an anthropomorphic hand with an optimized underactuated mechanism performs better grasps.

In recent years, using autonomous underwater vehicles (AUVs) for submarine missions has increased substantially. One of the problems in controlling these nonlinear devices is the possibility of a fault in the system operators. Failure causes increased operating costs and reduced vehicle performance. Therefore, the use of fault tolerance control is essential to ensure the stability and ability of the device to continue its activity. The focus of this article is on the trajectory tracking control for an underactuated AUV with actuator faults using kinematics and dynamics modeling. An adaptive rule is used as an online estimation to compensate for malfunctions in robot performance. In this regard, the adaptive fault-tolerant control plan is proposed, so that the closed-loop system is stable, and all control objectives are achievable. At first, the dynamic model of AUV with actuator fault and disturbance is described. Next, the control algorithm is designed for trajectory tracking in the presence of time-varying disturbances and actuator faults. The proposed adaptive rules will overcome disturbances and actuator faults. Finally, to illustrate the effectiveness of the proposed method, the provided controller is compared with other common control methods.

This paper presents a hierarchical framework that allows online point-to-point dynamic-stability-constrained optimal trajectory planning of a mobile manipulator robot working on rough terrain. First, the kinematics model of a mobile manipulator robot and the zero moment point stability measure are presented as theoretical background. Then, a sampling-based quasi-static planning algorithm modified for stability guarantee and traction optimization in continuous dynamic motion is presented along with a mathematical proof. The robot’s quasi-static path is then used as an initial guess to warm start a nonlinear optimal control solver which may otherwise have difficulties finding a solution to the stability-constrained formulation efficiently. The performance and computational efficiency of the framework are demonstrated through an application to a simulated timber harvesting mobile manipulator machine working on varying terrain. The results demonstrate feasibility of online trajectory planning on varying terrain while satisfying the dynamic stability constraint. Qualitative and quantitative comparisons with existing methods are also presented.

Body posture determination methods have many applications, including product design, ergonomic workplace design, human body simulation, virtual reality, and animation industry. Initiated in robotics, inverse kinematic (IK) method has been widely applied to proactive human body posture estimation. The analytic inverse kinematic (AIK) method is a convenient and time-saving type of IK methods. It is also indicated that, based on AIK methods, a specific body posture can be determined by the optimization of an arbitrary objective function. The objective of this paper is to predict the postures of human arms during reaching tasks. In this research, a human body model is established in MATLAB, where the middle rotation axis analytic kinematic method is accomplished, based on this model. The joint displacement function and joint discomfort function are selected to be initially applied in this AIK method. Results show that neither the joint displacement function nor the joint discomfort function predicts postures that are close enough to natural upper limb postures of human being, during reaching tasks. Therefore, a bi-criterion objective function is proposed by integrating the joint displacement function and joint discomfort function. The accuracy of the arm postures, predicted by the proposed objective function, is the most satisfactory, while the optimal value of the coefficient, in the proposed objective function, is determined by golden section search.

This article presents a methodology to reduce the energy consumption of an industrial robot. We propose a design for a 3R serial manipulator of general geometry. We show an analytical model aiming to analyze the search space of architectures based on the torsion angles of the robot to determine the optimal architecture that allows the efficient use of energy. The analytical model provides a theoretical estimation of the energy consumption and is validated by monitoring the experimental robot. The numerical calculations obtained with a particular case reduced the energy consumption by approximately 7.5%.

Both prosthetic and robotic research communities have tended to focus on hand/gripper development. However, the wrist unit could enable higher mobility of the end effector and thus more efficient and dexterous manipulation. The current state of the art in both prosthetic and robotic wrists is reviewed systematically, mainly concerning their kinematic structures and resultant capabilities. Further, by considering the biomechanical advantages of the human wrist, an evaluation including the mobility, stability, output capability, load capacity and flexibility of the current artificial wrists is conducted. With the pentagonal capability radar charts, the major limitations and challenges in the current development of artificial wrists are derived. This paper hence provides some useful insights for better robotic wrist design and development.