In this paper, we propose and implement an advanced manipulation framework that enables parametric learning of complex action trajectories along with their haptic feedback profiles. Our framework extends Dynamic Movement Primitives (DMPs) method with a new parametric nonlinear shaping function and a novel force-feedback coupling term. The nonlinear trajectories of the action control variables and the haptic feedback trajectories measured during execution are encoded with parametric temporal probabilistic models, namely parametric hidden Markov models (PHMMs). PHMMs enable autonomous segmentation of a taught skill based on the statistical information extracted from multiple demonstrations, and learning the relations between the model parameters and the properties extracted from the environment. Hidden states with high-variances in observation probabilities are interpreted as parts of the skill that could not be reliably learned and autonomously executed due to possibly uncertain or missing information about the environment. In those parts, our proposed force-feedback coupling term, which computes the deviation of the actual force feedback from the one predicted by the force-feedback PHMM, acts as a compliance term, enabling a human to scaffold the ongoing movement trajectory to accomplish the task. Our method is verified in a number of tasks including a real pick and place task that involves obstacles of different heights. Our robot, Baxter, successfully learned to generate the trajectory taking into the heights of the obstacles, move its end effector stiffly (and accurately) along the generated trajectory while passing through apertures, and allow human–robot collaboration in the autonomously detected segments of the motion, for example, when the gripper picks up the object whose position is not provided to the robot.

In this research, we propose a two-level control strategy for simultaneous gait generation and stable control of planar walking of the Assume The Robot Is A Sphere (ATRIAS) biped robot with unlocked torso, utilizing active spring-loaded inverted pendulum (ASLIP) as reference models. The upper level consists of an energy-regulating control calculated using the ASLIP model, producing reference ground reaction forces (GRFs) for the desired gait. In the lower level controller, PID force controllers for the motors ensure tracking of the reference GRFs for ATRIAS direct dynamics. Meanwhile, ATRIAS torso angle is controlled stably to make it able to follow a point mass template model. Advantages of the proposed control strategy include simplicity and efficiency. Simulation results using ATRIAS’s complete dynamic model show that the proposed two-level controller can reject initial condition disturbances while generating stable and steady walking motion.

In this paper, two strategies are proposed to optimize the energy consumption of a new screw in-pipe inspection robot which is steerable. In the first method, optimization is performed using the optimal path planning and implementing the Hamilton–Jacobi–Bellman (HJB) method. Since the number of actuators is more than the number of degrees of freedom of the system for the proposed steerable case, it is possible to minimize the energy consumption by the aid of the dynamics of the system. In the second method, the mechanics of the robot is modified by installing some turbine blades through which the drag force of the pipeline fluid can be employed to decrease the required propulsion force of the robot. It is shown that using both of the mentioned improvements, that is, using HJB formulation for the steerable robot and installing the turbine blades can significantly save power and energy. However, it will be shown that for the latter case this improvement is extremely dependent on the alignment of the fluid stream direction with respect to the direction of the robot velocity, while this optimization is independent of this case for the former strategy. On the other hand, the path planning dictates a special pattern of speed functionality while for the robot equipped by blades, saving the energy is possible for any desired input path. The correctness of the modeling is verified by comparing the results of MATLAB and ADAMS, while the efficiency of the proposed optimization algorithms is checked by the aid of some analytic and comparative simulations.

In this paper, we propose a novel unified framework for virtual guides. The human–robot interaction is based on a virtual robot, which is controlled by the admittance control. The unified framework combines virtual guides, control of the dynamic behavior, and path tracking. Different virtual guides and active constraints can be realized by using dead-zones in the position part of the admittance controller. The proposed algorithm can act in a changing task space and allows selection of the tasks-space and redundant degrees-of-freedom during the task execution. The admittance control algorithm can be implemented either on a velocity or on acceleration level. The proposed framework has been validated by an experiment on a KUKA LWR robot performing the Buzz-Wire task.

In this paper, we present and implement a novel approach for position-based visual servoing. The challenge of controlling the mobile robot while simultaneously estimating the camera to mobile robot transformation is solved. This is achieved using gradient descent (GD)-based estimation and the sliding-mode approach. The GD approach allows online parameter estimation for controlling the robot to achieve a desired position and orientation. The adaptive nature of the parameters demonstrates the robustness of the system. In contrast to existing work, the proposed technique achieves both estimation and control tasks in a single experiment. Simulation and experimental results are provided to validate the performance of the proposed scheme.

This paper investigates the vibrations of hexarot simulators. The generalized modeling of kinematics and dynamics formulation of a hexarot mechanism is addressed. This model considers the flexible manipulator with the base motion. The dynamic formulation has been developed based on the principle of virtual work. The dynamic model consists of the stiffness of the different parts of the mechanism, the effects of gravity and inertia, torque and force related to the joints viscous friction. Finally, the response of the end effector at various frequencies has been presented, and the vibrations of the mechanism and the dynamic stability index have been investigated.

Spray-painting equipments are important for the automatic spraying of long conical objects such as rocket fairing. This paper proposes a spray-painting equipment that consists of a feed worktable, a gantry frame and two serial–parallel mechanisms and investigates the optimal design of PRR–PRR parallel manipulator in serial–parallel mechanisms. Based on the kinematic model of the parallel manipulator, the conditioning performance, workspace and accuracy performance indices are defined. The dynamic model is derived using virtual work principle and dynamic evaluation index is defined. The conditioning performance, workspace, accuracy performance and dynamic performance are involved in multi-objective optimization design to determine the optimal geometrical parameters of the parallel manipulator. Furthermore, the geometrical parameters of the gantry frame are optimized. An example is given to show how to determine these parameters by taking a long object with conical surface as painted object.

A general serial–parallel manipulator connected in series by two different parallel manipulators with linear active legs is constructed. Its precise stiffness and elastic deformations are studied systematically. Its unified precise stiffness and precise elastic deformation models are established by considering both the moving links inertial wrench and the dynamic active/constrained wrench. A 3SPR+3RPS-type serial–parallel manipulator is illustrated for solving its precise stiffness and precise elastic deformation. The derived formulae of the precise stiffness and the precise elastic deformations of the general serial–parallel manipulator are verified by the theoretical solutions of the 3SPR+3RPS serial–parallel manipulator.

During visual servoing space activities, the attitude of free-floating space robot may be disturbed due to dynamics coupling between the satellite base and the manipulator. And the disturbance may cause communication interruption between space robot and control center on earth. However, it often happens that the redundancy of manipulator is not enough to fully eliminate this disturbance. In this paper, a method named off-line optimizing visual servoing algorithm is innovatively proposed to minimize the base disturbance during the visual servoing process where the degrees-of-freedom of the manipulator is not enough for a zero-reaction control. Based on the characteristic of visual servoing process and the robot system modeling, the optimal control method is applied to achieve the optimization, and a pose planning method is presented to achieve a second-order continuity of quaternion getting rid of the interruption caused by ambiguity. Then simulations are carried out to verify the method, and the results show that the robot is controlled with optimized results during visual servoing process and the joint trajectories are smooth.

The study of dexterous manipulation has provided important insights into human sensorimotor control as well as inspiration for manipulation strategies in robotic hands. Previous work focused on experimental environment with restrictions. Here, we describe a method using the deformation and color distribution of the fingernail and its surrounding skin to estimate the fingertip forces, torques, and contact surface curvatures for various objects, including the shape and material of the contact surfaces and the weight of the objects. The proposed method circumvents limitations associated with sensorized objects, gloves, or fixed contact surface type. In addition, compared with previous single finger estimation in an experimental environment, we extend the approach to multiple finger force estimation, which can be used for applications such as human grasping analysis. Four algorithms are used, c.q., Gaussian process, convolutional neural networks, neural networks with fast dropout, and recurrent neural networks with fast dropout, to model a mapping from images to the corresponding labels. The results further show that the proposed method has high accuracy to predict force, torque, and contact surface.

This paper proposes an adaptive robust impedance control for a single-link flexible arm when it encounters an environment at an unknown intermediate point. First, the intermediate collision point is estimated using a collision detection algorithm. The controller, then, switches from free to constrained motion mode. In the unconstrained motion mode, the exerted force to environment is nearly zero. Thus, the reference trajectory is a prescribed desired trajectory in position control. In the constrained motion mode, the reference trajectory is determined by the desired target dynamic impedance. The simulation results demonstrate the efficiency of proposed control scheme.

A design and manufacturing method is described for creating a motor tendon–actuated soft foam robot. The method uses a castable, light, and easily compressible open-cell polyurethane foam, producing a structure capable of large (~70% strain) deformations while requiring low torques to operate (<0.2 N·m). The soft robot can change shape, by compressing and folding, allowing for complex locomotion with only two actuators. Achievable motions include forward locomotion at 13 mm/s (4.3% of body length per second), turning at 9◦/s, and end-over-end flipping. Hard components, such as motors, are loosely sutured into cavities after molding. This reduces unwanted stiffening of the soft body. This work is the first demonstration of a soft open-cell foam robot locomoting with motor tendon actuators. The manufacturing method is rapid (~30 min per mold), inexpensive (under $3 per robot for the structural foam), and flexible, and will allow a variety of soft foam robotic devices to be produced.

The paper reports on a new multi-view algorithm that combines information from multiple images of a single target object, captured at different distances, to determine the identity of an object. Due to the use of global feature descriptors, the method does not involve image segmentation. The performance of the algorithm has been evaluated on a binary classification problem for a data set consisting of a series of underwater images.

This paper studies the stabilization problem for a class of underactuated systems in the presence of unknown disturbances. Due to less number of control inputs with respect to the degrees of freedom of the system, closed-loop asymptotic stability is a challenging issue in this field. In this paper, anti-swing controllers are designed for nominal and disturbed systems. In the case of the nominal system, the proposed two-loop controller is a combination of collocated partial feedback linearization and hierarchical sliding mode control (HSMC) theories. Then, due to the importance of robustness in control of physical systems, the proposed controller is developed for underactuated mechanical systems in the presence of additive disturbances. One of the main advantages of the proposed design method is that it does not need any switching algorithm. Finally, to illustrate the performance of the proposed controllers, they are applied to two underactuated mechanical systems: a pendubot and a Furuta pendulum. In addition, the practicality of the proposed approach is also verified experimentally using a quadrotor stand.

The progressive automation framework allows the seamless transition of a robot from kinesthetic guidance to autonomous operation mode during programming by demonstration of discrete motion tasks. This is achieved by the synergetic action of dynamic movement primitives (DMPs), virtual fixtures, and variable impedance control. The proposed DMPs encode the demonstrated trajectory and synchronize with the current demonstration from the user so that the reference generated motion follows the human’s demonstration. The proposed virtual fixtures assist the user in repeating the learned kinematic behavior but allow penetration so that the user can make modifications to the learned trajectory if needed. The tracking error in combination with the interaction forces and torques is used by a variable stiffness strategy to adjust the progressive automation level and transition the leading role between the human and the robot. An energy tank approach is utilized to apply the designed controller and to prove the passivity of the overall control method. An experimental evaluation of the proposed framework is presented for a pick and place task and results show that the transition to autonomous mode is achieved in few demonstrations.

In this paper we address the problem of coverage path planning (CPP) for multiple cooperating mobile robots. We use a ‘partition and cover’ approach using Voronoi partition to achieve natural passive cooperation between robots to avoid task duplicity. We combine two generalizations of Voronoi partition, namely geodesic-distance-based Voronoi partition and Manhattan-distance-based Voronoi partition, to address contiguity of partition in the presence of obstacles and to avoid partition-boundary-induced coverage gap. The region is divided into 2D×2D grids, where D is the size of the robot footprint. Individual robots can use any of the single-robot CPP algorithms. We show that with the proposed Geodesic-Manhattan Voronoi-partition-based coverage (GM-VPC), a complete and non-overlapping coverage can be achieved at grid level provided that the underlying single-robot CPP algorithm has similar property.We demonstrated using two representative single-robot coverage strategies, namely Boustrophedon-decomposition-based coverage and Spanning Tree coverage, first based on so-called exact cellular decomposition and second based on approximate cellular decomposition, that the proposed partitioning scheme completely eliminates coverage gaps and coverage overlaps. Simulation experiments using Matlab and V-rep robot simulator and experiments with Fire Bird V mobile robot are carried out to validate the proposed coverage strategy.

This paper proposes a novel control framework for a single-master/multi-slave teleoperation system to grasp and handle an object, considering nonlinearity and uncertainty in the dynamics of the slaves and time-varying delay in the communication channel. Based on passive decomposition approach, the dynamics of the slaves are decomposed into two decoupled systems, and then two higher order sliding mode controllers are designed to control them. Also, a synchronization control methodology for the master is developed. Stability is fully studied using the passivity property and Lyapunov theorem. Finally, simulation and practical results confirm that the control system works well against the conditions.

Joint flexibility introduces additional degrees of freedom and vibration modes, thus limiting the performance of the manipulator. To improve control bandwidth, this paper proposes an enhanced two-stage state feedback (SFB) controller, which combines two parts. The first is a SFB loop, which considers the motor position as a virtual control input for the link side dynamics. The second is a disturbance-state observer, which compensates disturbances and reconstructs indirect measurements. Experimental results show the effectiveness of the proposed controller in terms of position tracking, link vibration, and rejection of the kinematic error from the joint’s harmonic drive reducer.

This paper tackles the challenge of the necessity of using the sequence of past environment states as the controller’s inputs in a vision-based robot navigation task. In this task, a robot has to follow a given trajectory without falling in pits and missing its balance in uneven terrain, when the only sensory input is the raw image captured by a camera. The robot should distinguish big pits from small holes to decide between avoiding and passing over. In non-Markov processes such as the abovementioned task, the decision is done using past sensory data to ensure admissible performance. Applying images as sensory inputs naturally causes the curse of dimensionality difficulty. On the other hand, using sequences of past images intensifies this difficulty. In this paper, a new framework called recurrent deep learning (RDL) with combination of deep learning (DL) and recurrent neural network is proposed to cope with the above challenge. At first, the proper features are extracted from the raw image using DL. Then, these represented features plus some expert-defined features are used as the inputs of a fully connected recurrent network (as target network) to generate command control of the robot. To evaluate the proposed RDL framework, some experiments are established on WEBOTS and MATLAB co-simulation platform. The simulation results demonstrate the proposed framework outperforms the conventional controller based on DL for the navigation task in the uneven terrains.

This paper studies the gait characteristics of a quadruped rover that mimics domestic cats, and attempts to optimize these characteristics. The kinematics and dynamics formulation of the rover’s three-dimensional model is developed, and its gait, pose and corresponding control parameters are computed to minimize torque or maximize speed, using a genetic algorithm. The optimization model consists of a set of equality and inequality constraints that ensure the feasibility and stability of the gaits, while considering the entire gait spectrum that feline species exhibit. The optimal gaits for minimizing the torque closely resemble lateral sequence gaiting, with a trotting behaviour as speed increases. A running gait is obtained at the maximum speed. The optimization results appear to conform to the biological observations of feline species, suggesting the tendency of conserving energy in biological gaiting.