FastSLAM 2.0 is a popular framework which uses a Rao-Blackwellized particle filter to solve the simultaneous localization and mapping problem. The sampling process is one of the most important phases in the FastSLAM 2.0 framework. Its estimation accuracy depends heavily on a correct prior knowledge about the control and observation noise statistics (the covariance matrices Q and R). Without the correct prior knowledge about these matrices, the estimation accuracy of the robot path and landmark positions may degrade seriously. However in many applications, the prior knowledge is unknown, or these noises are non-stationary. In this paper, these covariance matrices are supposed to be dynamic and denoted as Qt and Rt. Since there are noises, time-adjacent observations are inconsistent with each other. This inconsistency can reflect the real value of the covariance matrices. By the inconsistency, an extra step is introduced to the FastSLAM 2.0 framework. This step makes Qt and Rt match with their real value by using a particle swarm optimization method based on fractional calculus and alpha stable distribution (FC&ASD-PSO). Both simulation and experimental results show that the proposed algorithm improves the accuracy by the more accurate estimation on the noise covariance matrices.

Several non-holonomic Dubins-car-like robots travel over paths with bounded curvatures in a plane that contains an a priori unknown region. The robots are anonymous to one another and do not use communication facilities. Any of them has access to the current minimum distance to the region and can determine the relative positions and orientations of the other robots within a finite and given visibility range. We present a distributed navigation and guidance strategy under which every robot autonomously converges to the desired minimum distance to the region with always respecting a given safety margin, the robots do not collide with one another and do not get into clusters, and the entire team ultimately sweeps over the respective equidistant curve at a speed exceeding a given threshold, thus forming a kind of a sweeping barrier at the perimeter of the region. Moreover, this strategy provides effective sub-uniform distribution of the robots over the equidistant curve. Mathematically rigorous justification of the proposed strategy is offered; its effectiveness is confirmed by extensive computer simulations and experiments with real wheeled robots.

A new motion-free iterative closest point (ICP) algorithm is proposed for building a two-dimensional (2D) map for mobile robot navigation. A laser range finder (LRF) sensor is installed on a mobile robot to scan and measure the depth data of the environment to form a 2D map during mobile robot navigation. Because the scanning and navigation motions are performed independently, the scanned data contain distortions from the motions of the mobile robot. To compensate for the distortions, the proposed motion-free ICP algorithm estimates the effects of the dynamic motions of the robot on the scanning process. That is, the motion-free algorithm compensates for the distance measurement errors related to the dynamic changes in the mobile robot's velocity. Experiments were performed with actual velocity changes of a mobile robot to demonstrate and verify the effective performance of the proposed algorithm.

We propose a novel stereo visual IMU-assisted (Inertial Measurement Unit) technique that extends to large inter-frame motion the use of KLT tracker (Kanade–Lucas–Tomasi). The constrained and coherent inter-frame motion acquired from the IMU is applied to detected features through homogenous transform using 3D geometry and stereoscopy properties. This predicts efficiently the projection of the optical flow in subsequent images. Accurate adaptive tracking windows limit tracking areas resulting in a minimum of lost features and also prevent tracking of dynamic objects. This new feature tracking approach is adopted as part of a fast and robust visual odometry algorithm based on double dogleg trust region method. Comparisons with gyro-aided KLT and variants approaches show that our technique is able to maintain minimum loss of features and low computational cost even on image sequences presenting important scale change. Visual odometry solution based on this IMU-assisted KLT gives more accurate result than INS/GPS solution for trajectory generation in certain context.

When the end-effector of a robotic arm grasps different payload masses, the output of joint motion will vary. By using a model reference adaptive control approach, the payload variation effect can be solved. This paper describes the design for a hybrid controller for serial robotic manipulators by combining a PID controller and a model reference adaptive controller (MRAC) in order to further improve the accuracy and joint convergence speed performance. The convergence performance of the PID controller, the MRAC and the PID+MRAC hybrid controller for 1-DOF, 2-DOF and subsequently 3-DOF manipulators is compared. The comparison results show that the convergence speed and its performance for the MRAC and the PID+ MRAC controllers is better than that of the PID controller, and the convergence performance for the hybrid control is better than that of the MRAC control.

A multimodal actuator was proposed to achieve a more agile power-assisted exoskeleton in uncertain complex walking environments. A power-assisted knee exoskeleton prototype based on a multimodal actuator was constructed. With this multimodal actuator, several modes of operation in the power-assisted knee exoskeleton during a motion cycle are actuated, including series elastic actuation, stiff position control, and energy storage and release. Also, a control strategy for power-assisted knee exoskeleton motion control based on a state machine is developed. The ability of the power-assisted knee exoskeleton to follow human motion was tested, and the results showed that the angle error of the knee exoskeleton followed the human motion is not more than 0.4˚, and the response time error of the knee exoskeleton followed the human motion is not more than 0.2 s.

Most control algorithms for rigid-link electrically driven robots are given in joint coordinates. However, since the task to be accomplished is expressed in Cartesian coordinates, inverse kinematics has to be computed in order to implement the control law. Alternatively, one can develop the necessary theory directly in workspace coordinates. This has the disadvantage of a more complex robot model. In this paper, a robust control scheme is given to achieve exact Cartesian tracking without the knowledge of the manipulator kinematics and dynamics, actuator dynamics and nor computing inverse kinematics. The control design procedure is based on a new form of universal approximation theory and using Stone–Weierstrass theorem, to mitigate structured and unstructured uncertainties associated with external disturbances and actuated manipulator dynamics. It has been assumed that the lumped uncertainty can be modeled by linear differential equations. As the method is Model-Free, a broad range of manipulators can be controlled. Numerical case studies are developed for an industrial robot manipulator.