Performing actions in a timely manner is an indispensable aspect in everyday human activities. Accordingly, it has to be present in robotic systems if they are going to seamlessly interact with humans. The current work addresses the problem of learning both the spatial and temporal characteristics of human motions from observation. We formulate learning as a mapping between two worlds (the observed and the action ones). This mapping is realized via an abstract intermediate representation termed “Latent Space.” Learned actions can be subsequently invoked in the context of more complex human–robot interaction (HRI) scenarios. Unlike previous learning from demonstration (LfD) methods that cope only with the spatial features of an action, the formulated scheme effectively encompasses spatial and temporal aspects. Learned actions are reproduced under the high-level control of a time-informed task planner. During the implementation of the studied scenarios, temporal and physical constraints may impose speed adaptations in the reproduced actions. The employed latent space representation readily supports such variations, giving rise to novel actions in the temporal domain. Experimental results demonstrate the effectiveness of the proposed scheme in the implementation of HRI scenarios. Finally, a set of well-defined evaluation metrics are introduced to assess the validity of the proposed approach considering the temporal and spatial consistency of the reproduced behaviors.

Mission planning is a complex motion planning problem specified by using Temporal Logic constituting of Boolean and temporal operators, typically solved by model verification algorithms with an exponential complexity. The paper proposes co-evolutionary optimization thus building an iterative solution to the problem. The language for mission specification is generic enough to represent everyday missions, while specific enough to design heuristics. The mission is broken into components which cooperate with each other. The experiments confirm that the robot is able to outperform the search, evolutionary and model verification techniques. The results are demonstrated by using a Pioneer LX robot.

Conventional simultaneous localization and mapping (SLAM) has concentrated on two-dimensional (2D) map building. To adapt it to urgent search and rescue (SAR) environments, it is necessary to combine the fast and simple global 2D SLAM and three-dimensional (3D) objects of interest (OOIs) local sub-maps. The main novelty of the present work is a method for 3D OOI reconstruction based on a 2D map, thereby retaining the fast performances of the latter. A theory is established that is adapted to a SAR environment, including the object identification, exploration area coverage (AC), and loop closure detection of revisited spots. Proposed for the first is image optical flow calculation with a 2D/3D fusion method and RGB-D (red, green, blue + depth) transformation based on Joblove–Greenberg mathematics and OpenCV processing. The mathematical theories of optical flow calculation and wavelet transformation are used for the first time to solve the robotic SAR SLAM problem. The present contributions indicate two aspects: (i) mobile robots depend on planar distance estimation to build 2D maps quickly and to provide SAR exploration AC; (ii) 3D OOIs are reconstructed using the proposed innovative methods of RGB-D iterative closest points (RGB-ICPs) and 2D/3D principle of wavelet transformation. Different mobile robots are used to conduct indoor and outdoor SAR SLAM. Both the SLAM and the SAR OOIs detection are implemented by simulations and ground-truth experiments, which provide strong evidence for the proposed 2D/3D reconstruction SAR SLAM approaches adapted to post-disaster environments.

We present a robot navigation system based on Behavioral Finite State Social Machine. The paper makes a robot operate as a social tour guide that adapts its navigation based on the behavior of the visitors. The problem of a robot leading a human group with a limited field-of-view vision is relatively untouched in the literature. Uncertainties arise when the visitors are not visible, wherein the behavior of the robot is adapted as a social response. Artificial potential field is used for local planning, and a velocity manager sets the speed disproportional to time duration of missing visitors.

In this paper, a general solution to the formation control problem without collisions for first-order multi-agent systems is proposed. The case of an arbitrary number of mobile agents on a plane with saturated input velocity is analysed. Besides, conditions on the communication graph among agents are relaxed to the only requirement of containing a directed spanning tree. This general approach is an extended result from the simpler case of combinations of cyclic pursuit communication graphs. The proposed solution to this problem is designed in two steps. First, the asymptotic convergence in the absence of collisions is ensured. After this, the non-collision problem is faced by analysing the most general possible geometrical scenario which can lead to collision among agents. Discontinuous vector fields with unstable counterclockwise focus behaviour are applied by every agent in order to repel each other. Numerical simulations illustrate the performance of the proposed scheme.

Stability is a major problem with bilateral teleoperator when there is a transmission time delay. By adding adequate damping gain into position servo, a stable bilateral control can be achieved under the existence of time delay. However, large damping gain is required to stabilize the system and the performance is degraded. In this paper, the performance improvement by introducing a high-pass filter into a proportional derivative-based teleoperator with time delay has been studied. We proposed new control law and derived the stability condition. We demonstrated the performance improvement using a hybrid matrix, 1-degree of freedom (DOF), 2-DOF simulations, and 2-DOF peg-in-hole experiments.

The problem of hybrid force and motion control over unknown rigid surfaces when only joint position measurements are available is considered. To overcome this problem, an extended state high-gain observer is designed to simultaneously estimate the contact force and joint velocities. These estimated signals are in turn employed to design a local estimator of the unknown surface gradient. This gradient is utilized to decompose the task space into two orthogonal subspaces: one for force tracking and the other one for motion control. A simple position Proportional Integral Derivative (PID) and force Proportional Integral (PI) controllers are proposed to track the desired signals. Finally, a mathematical analysis of the closed-loop dynamics is carried out, guaranteeing uniform ultimate boundedness of the position and force tracking errors and of the surface gradient estimation error. A numerical simulation is employed to validate the approach in an ideal scenario, while experiments are carried out to test the proposed strategy when uncertainties and unmodeled dynamics are present.

Biological fish can create high forward swimming speed due to change of thrust/drag area of pectoral fins between power stroke and recovery stroke in rowing mode. In this paper, we proposed a novel type of folding pectoral fins for the fish robot, which provides a simple approach in generating effective thrust only through one degree of freedom of fin actuator. Its structure consists of two elemental fin panels for each pectoral fin that connects to a hinge base through the flexible joints. The Morison force model is adopted to discover the relationship of the dynamic interaction between fin panels and surrounding fluid. An experimental platform for the robot motion using the pectoral fin with different flexible joints was built to validate the proposed design. The results express that the performance of swimming velocity and turning radius of the robot are enhanced effectively. The forward swimming velocity can reach 0.231 m/s (0.58 BL/s) at the frequency near 0.75 Hz. By comparison, we found an accord between the proposed dynamic model and the experimental behavior of the robot. The attained results can be used to design controllers and optimize performances of the robot propelled by the folding pectoral fins.

This paper presents a soft robot which can imitate the crawling locomotion of an earthworm. Locomotion of the robot can be achieved by expanding and contracting the body that is made of flexible material. A link of the earthworm-like robot is combined with three modules, and a multi-cavity earthworm-like soft robot is combined with multiple links. The multiple links of the earthworm-like soft robot are fabricated by silicone in the three-dimensional printed customized molds. Experiments on a single module, two-links, and three-links show that the soft robot can move and bend on condition of modules extension and contraction in a specified gait. The development of the earthworm-like soft robot shows a great prospect in many complicated environments such as pipeline detection.

The Inverse Kinematics (IK) problem of manipulators can be divided into two distinct steps: (1) Problem formulation, where the problem is developed into a form which can then be solved using various methods. (2) Problem solution, where the IK problem is actually solved by producing the values of different joint space variables (joint angles, joint velocities or joint accelerations). The main focus of this paper is concentrated on the discussion of the IK problem of redundant manipulators, formulated as a quadratic programming optimization problem solved by different kinds of recurrent neural networks.

In this paper, optimal control of a 3PRS robot is performed, and its related optimal path is extracted accordingly. This robot is a kind of parallel spatial robot with six DOFs which can be controlled using three active prismatic joints and three passive rotary ones. Carrying a load between two initial and final positions is the main application of this robot. Therefore, extracting the optimal path is a valuable study for maximizing the load capacity of the robot. First of all, the complete kinematic and kinetic modeling of the robot is extracted to control and optimize the robot. As the robot is categorized as a constrained robot, its kinematics is studied using a Jacobian matrix and its pseudo inverse whereas its kinetics is studied using Lagrange multipliers. The robot is then controlled using feedforward term of the inverse dynamics. Afterward, the extracted dynamics equation of the robot is transferred to state space to be employed for calculus of variations. Considering the constrained entity of the robot, null space of the robot is employed to eliminate the Lagrange multipliers to provide the applicability of indirect variation algorithm for the robot. As a result, not only are the optimal controlling signals calculated but also the corresponding optimal path of the robot between two boundary conditions is extracted. All the modeling, controlling, and optimization process are verified using MATLAB simulation. The profiles are then double-checked by comparing the results with SimMechanics. It is proved that with the aid of the proposed controlling and optimization method of this article, the robot can be controlled along its optimal path through which the maximum load can be carried.

In order to make the end of the three-axis platform follow the control command and achieve stable control of the end attitude, an improved orientation vector spherical linear interpolation (SLERP) method is proposed for the requirements, which specifically handles the position of the gimbal lock, so that the platform can move smoothly around the gimbal lock position. A three-axis platform with a camera at the end is set up for the validity of the proposed algorithm. At first, an adaptive speed measurement method based on incremental encoder is introduced, which can automatically adapt to high and low speed, and estimate the ultra-low speed to realize the speed measurement of large dynamic range, and this is used for the motion control of the three-axis platform. Then, the SLERP method for the quaternion interpolation on the starting and ending attitudes represented in quaternion is introduced in detail, and it is continuously improved in response to its existing problems for the platform. Finally, an orientation vector SLERP method is proposed, which uses viscosity factor and rejection factor to adjust the algorithm near the platform’s gimbal lock position. A tracking experiment was designed using the red ball as the following target detected by the designed target tracking algorithm using the camera, which verified the effectiveness of the attitude tracking control based on the proposed improved orientation vector SLERP.

This paper deals with the problem of navigating semi-autonomous mobile robots without global localization systems in unknown environments. We propose a planning-based obstacle avoidance strategy that relies on local maps and a series of short-time coordinate frames. With this approach, simple odometry and range information are sufficient to make the robot to safely follow the user commands. Different from reactive obstacle avoidance strategies, the proposed approach chooses a good and smooth local path for the robot. The methodology is evaluated using a mobile service robot moving in an unknown corridor environment populated with obstacles and people.

This paper documents autonomous multi-floor stairwell ascent by a legged robot. This is made possible through empirically deployed sequential composition of several reactive controllers, with perceptually triggered transitions. This composition relies on simplified assumptions regarding the robot’s sensory capabilities, its level of mobility, and the environment it operates in. The discrepancies between these assumptions and the physical reality are capably handled by the intrinsic motor competence of the robot. This behavior is implemented on the legged RHex platform and experiments spanning 10 different stairwells with various challenges are conducted.

The enhanced dexterity and manipulability offered by master–slave teleoperated surgical systems have significantly improved the performance and safety of minimally invasive surgeries. However, effective manipulation of surgical robots is sometimes limited due to the mismatch between the slave and master robots’ kinematics and workspace. The purpose of this paper is first to formulate a quantifiable measure of the combined master–slave system manipulability. Next, we develop a null-space controller for the redundant master robot that employs the proposed manipulability index to enhance the performance of teleoperation tasks by matching the kinematics of the redundant master robot with the kinematics of the slave robot. The null-space controller modulates the redundant degrees of freedom of the master robot to reshape its manipulability ellipsoid (ME) towards the ME of the slave robot. The ME is the geometric interpretation of the kinematics of a robot. By reshaping the master robot’s manipulability, we match the master and slave robots’ kinematics. We demonstrate that by using a redundant master robot, we are able to enhance the master–slave system manipulability and more intuitively transfer the slave robot’s dexterity to the user. Simulation and experimental studies are performed to validate the performance of the proposed control strategy. Results demonstrate that by employing the proposed manipulability index, we can enhance the user’s control over the force/velocity of a surgical robot and minimize the user’s control effort for a teleoperated task.

In this paper, a multi-objective design optimization of the 3-UPU translational parallel manipulator is presented. Based on a new algorithm, which combines the genetic algorithms and the Krawczyk operator, the robot position error is minimized and the robot design parameters tolerances are maximized, simultaneously. The results show that the designer can maintain the manipulator accuracy by using a specific size of the base, and can restrict its tolerance even by enlarging the actuators’ tolerance intervals. This algorithm is also used to determine the maximum design parameters tolerances for an allowable robot position error. The proposed algorithm can be extended to optimize other types of robots.

This paper presents a neural network-based four-direction search scheme of path planning for mobile agents, given a known environmental map with stationary obstacles. Firstly, the map collision energy is modeled for all the obstacles based on neural network. Secondly, for the shorted path-search purpose, the path energy is considered. Thirdly, to decrease the path-search time, a variable step-length is designed with respect to collision energy of the previous iteration path. Simulation results demonstrate that the variable step-length is effective and can decrease the iteration time substantially. Lastly, experimental results show that the mobile agent tracks the generated path well. Both the simulation and experiment results substantiate the feasibility and realizability of the presented scheme.