There are tasks that do not require a controlled motion in all spatial directions. These unused degrees-of-freedom (DOF) make the robot functionally redundant. Traditional methods for redundancy resolution developed for intrinsic redundant robots cannot be used directly for resolving functional redundancy. The functional redundancy is in general not reflected as rows in the Jacobian matrix and therefore, the unused DOFs cannot be mapped into the null-space. In the paper we present a novel approach for the orientation related functional redundancy resolution based on the task space rotation. We define a time-variant task space frame so that one or more frame axes are aligned with the unused directions. This allows that the rows of Jacobian matrix corresponding to functional redundant DOFs are excluded. As the standard quaternion control is not appropriate to control orientation of such functional redundant robots, we propose two control algorithms to control one or two orientation directions. Two examples then show how the proposed control algorithms can be implemented and finally, we present the application of the algorithm on a two-arm robot system.

The creation and adaptation of motor behaviors is an important capability for autonomous robots. In this paper we propose an approach for altering existing robot behaviors online, where a human coach interactively changes the robot motion to achieve the desired outcome. Using hand gestures, the human coach can specify the desired modifications to the previously acquired behavior. To preserve a natural posture while performing the task, the movement is encoded in the robot's joint space using periodic dynamic movement primitives. The coaching gestures are mapped to the robot joint space via robot Jacobian and used to create a virtual force field affecting the movement. A recursive least squares technique is used to modify the existing movement with respect to the virtual force field. The proposed approach was evaluated on a simulated three degrees of freedom planar robot and on a real humanoid robot, where human coaching gestures were captured by an RGB-D sensor. Although our focus was on rhythmic movements, the developed approach is also applicable to discrete (point-to-point) movements.

In this paper we address the problem of accurate trajectory tracking while ensuring compliant robotic behaviour for periodic tasks. We propose an approach for on-line learning of task-specific dynamics, i.e. task specific movement trajectories and corresponding force/torque profiles. The proposed control framework is a multi-step process, where in the first step a human tutor shows how to perform the desired periodic task. A state estimator based on an adaptive frequency oscillator combined with dynamic movement primitives is employed to extract movement trajectories. In the second step, the movement trajectory is accurately executed in the controlled environment under human supervision. In this step, the robot is accurately tracking the acquired movement trajectory, using high feedback gains to ensure accurate tracking. Thus it can learn the corresponding force/torque profiles, i. e. task-specific dynamics. Finally, in the third step, the movement is executed with the learned feedforward task-specific dynamic model, allowing for low position feedback gains, which implies compliant robot behaviour. Thus, it is safe for interaction with humans or the environment. The proposed approach was evaluated on a Kuka LRW robot performing object manipulation and crank turning.

The framework of dynamic movement primitives allows the generation of discrete and periodic trajectories, which can be modulated in various aspects. We propose and evaluate a novel modulation approach that includes force feedback and thus allows physical interaction with objects and the environment. The proposed approach also enables the coupling of independently executed robotic trajectories, simplifying the execution of bimanual and tightly coupled cooperative tasks. We apply an iterative learning control algorithm to learn a coupling term, which is applied to the original trajectory in a feed-forward fashion. The coupling term modifies the trajectory in accordance to either the desired position or external force. The strengths of the approach are shown in bimanual or two-agent obstacle avoidance tasks, where no higher level cognitive reasoning or planning are required. Results of simulated and real-world experiments on the ARMAR-III humanoid robot in interaction and object lifting tasks, and on two KUKA LWR robots in a bimanual setting are presented.

We address the problem of extracting the fundamental frequency of an arbitrary periodic or quasi-periodic signal for application in robotic tasks. We focus on controlling periodic robotic movement by extracting the frequency of human movement by using surface electromyography (EMG), a technique by which muscle action potentials are gathered by electrodes placed on the skin. However, since the EMG signal is quasi-periodic with a lot of frequency components and noise, it is difficult to determine the frequency and phase of the measured limb motion. We propose to use nonlinear dynamical systems capable of extracting the frequency and the phase from an unknown periodic signal with an arbitrary waveform. The method uses a whole Fourier series representation in a feedback loop. It is capable of extracting the frequency and the phase of an unknown periodic signal in real-time, without any additional signal processing or preprocessing. Combining this method with an output dynamic system based on dynamic movement primitives, which generate the desired trajectory, allows synchronization between human muscles actions and some other system actions like robot motion or electrical neuromuscular stimulator activation.

In this paper we propose a novel method for ensuring stability of a skiing robot on a previously unknown ski slope. The method is based on a null-space velocity control algorithm, which is essential for achieving good behavior of redundant robotic systems. Using the pseudo-inverse, the task and the null-space motions are decoupled. For the skiing robot, the primary task is stability, the secondary task is the direction and the third task is the pose of the robot. The dynamic stability index of the skiing robot is defined with the normalized zero moment point. We demonstrated the effectiveness of the proposed method both in simulation in the virtual reality environment as well as on the real robot.

In the paper we address a problem of controlling an oscillating motion with a robot. As the object we have selected a yo-yo. Yo-yo is a toy made of two discs connected with a short thin axle, which can be moved up and down by moving a string tied to the axle. Although it is very easy to play with yo-yo for a human, designing a control system for a robot is quite a challenge. We present a simplified model of a yo-yo, which is intended for control analysis and design. Our model has one degree-of-freedom and the behaviour at the end of the string is modelled as an impact. Next we discuss the control strategy and present a control method, which allows playing the yo-yo at a selected top height.

The paper deals with kinematic control algorithms for on-line obstacle avoidance which allow a kinematically redundant manipulator to move in an unstructured environment without colliding with obstacles. The presented approach is based on the redundancy resolution at the velocity level. The primary task is determined by the end-effector trajectories and for obstacle avoidance the internal motion of the manipulator is used. The obstacle avoiding motion is defined in one-dimensional operational space and, hence, the system has less singularities making implementation easier. Instead of the exact pseudoinverse solution we propose an approximate one which is computationally more efficient and allows us to consider many simultaneously active obstacles without any problems. The fast cycle times of the numerical implementation enable use of the algorithm in real-time control. For illustration, some simulation results of a highly redundant planar manipulator moving in an unstructured and time-varying environment and experimental results of a four link planar manipulator are given.

We deal with implementation of force control on a planar 4 DOF experimental redundant robot. A globally stable impedance force control law is used to control force and trajectory in the operational space. Redundancy is solved by a gradient optimization technique using dynamically consistent pseudo-inverse. In our case we implemented two optimization criteria-obstacle avoidance and torque optimization, important for force controlled redundant robots. The desired null space motion is controlled with a new formulation of the null space velocity controller, that assures globally stable behavior in null space. The performance of the proposed control law was tested on the experimental 4 DOF planar manipulator.

In the paper we present a solution to the optimal trajectory planning problem subject to constraints given by joint torques and velocities, and the task. The main result is a trajectory planning algorithm which enables the calculation of the optimal motion due to the given requirements and taking into account the dynamics of the manipulator. First, the dynamic model of the manipulator in the parametric form (as a function of the scalar path parameter) is given. Next, the constraints on torques and velocities are converted to those on the path parameter. Using these, the task and joint trajectories are calculated. In the end, some examples are given to illustrate the capabilities of the given trajectory planning algorithm.

The authors deal with the implementation of adaptive control of force-controlled manipulators. Due to the hardware and software limitations of existing industrial robots, the force control loop must be realized in many cases with an inner velocity and position controller. In such cases, the response of the robot becomes sensitive to the changes of the environment stiffness. To solve the problem, an adaptive controller has been proposed by many authors. The main drawback of these controllers is that they cannot assure satisfactory behavior in the case when both environment stiffness and environment/robot contact position are changed simultaneously. Therefore a dual force sensor is proposed that is passive and active. which allows estimation of the contact position and the environment stiffness simultaneously. Adaptive force control was tested on the deburring process of a workpiece.>