Sliding mode observer is a variable structure system where the dynamics of a nonlinear system is altered via application of a high-frequency switching control. This paper presents a non-linear sliding mode observer for wheel linear slip and slip angle estimation of a single wheel based on its kinematic model and velocity measurements with added noise to simulate actual on-board sensor measurements. Lyapunov stability theory is used to establish the stability conditions for the observer. It is shown that the observer will converge in a finite time, provided the observer gains satisfy constraints based on a stability analysis. To validate the observer, linear and two-dimensional (2D) test rigs are specially designed. The sliding mode observer is tested under a variety of conditions and it is shown that the sliding mode observer can estimate wheel slip and slip angle to a high accuracy. It is also shown that the sliding mode observer can accurately predict wheel slip and slip angle in the presence of noise, by testing the performance of the sliding mode observer after adding white noise to the measurements. An extended Kalman filter is also developed for comparison purposes. The sliding mode observer is better in terms of prediction accuracy.

This paper presents a novel microstep control approach for improving the positioning precision of a millimeter-sized omni-directional mobile microrobot. The microrobot that is designed for microassembly in a microfactory is driven by three electromagnetic micromotors. The structures of both the microrobot and the micromotor are described in the paper. The torque vectors synthesis method (TVSM) is developed and the torque self-balance principle (TSBP) is employed to realize the microstep control. The control circuit that needs a complex digital logic is devised with a complex programmable logic device (CPLD). The experiments using a hard disk and the real microrobot demonstrated the validity and performance of TVSM and TSBP. And the angular step precision of the microrobot is increased by three times compared to the normal control.

The aim of this paper is to determine the optimal trajectory and maximum payload of flexible link manipulators in point-to-point motion. The method starts with deriving the dynamic equations of flexible manipulators using combined Euler–Lagrange formulation and assumed modes method. Then the trajectory planning problem is defined as a general form of optimal control problem. The computational methods to solve this problem are classified as indirect and direct techniques. This work is based on the indirect solution of open-loop optimal control problem. Because of the offline nature of the method, many difficulties like system nonlinearities and all types of constraints can be catered for and implemented easily. By using the Pontryagin's minimum principle, the obtained optimality conditions lead to a standard form of a two-point boundary value problem solved by the available command in MATLAB®. In order to determine the optimal trajectory a computational algorithm is presented for a known payload and the other one is then developed to find the maximum payload trajectory. The optimal trajectory and corresponding input control obtained from this method can be used as a reference signal and feedforward command in control structure of flexible manipulators. In order to clarify the method, derivation of the equations for a planar two-link manipulator is presented in detail. A number of simulation tests are performed and optimal paths with minimum effort, minimum effort-speed, maximum payload, and minimum vibration are obtained. The obtained results illustrate the power and efficiency of the method to solve the different path planning problems and overcome the high nonlinearity nature of the problems.

In this paper, a biped that combines the merits of both powered and passive bipeds is proposed. The semi-activeness of the biped is due to the fact that during most of a walking cycle, only half of the joints are actuated to follow specific trajectories, and the other half of joints remain unactuated but have passive springs connected between adjacent links. It is expected that by having unactuated joints, the biped can preserve the power-saving feature of the passive biped, and by having actuated joints under active control, the biped can also achieve the stability and performance of the powered biped. To devise a systematic design methodology for the biped, its dynamics as well as the walking constraints are carefully studied. Furthermore, an optimization procedure is also proposed to compute the optimal trajectories for the actuated joints and spring constants, which can lead to minimum energy consumption. The feasibility of the proposed biped, including the system design and the control strategy, is verified by hardware implementation. Experiments indicate that the biped walking in the semi-active manner consumes 80% less the electrical power than the fully powered biped that performs the same gait and is more energy-efficient than several state-of-the-art bipeds.

FastSLAM is a framework for simultaneous localization and mapping using a Rao-Blackwellized particle filter (RBPF). But, FastSLAM is known to degenerate over time due to the loss of particle diversity, mainly caused by the particle depletion problem in resampling phase. In this work, improved particle filter using geometric relation between particles is proposed to restrain particle depletion and to reduce estimation errors and error variances. It uses a KD tree (k-dimensional tree) to derive geometric relation among particles and filters particles with importance weight conditions for resampling. Compared to the original particle filter used in FastSLAM, this technique showed less estimation error with lower error standard deviation in computer simulations.

Several problems of practical interest in robotics can be modelled as the convolution of functions on the Euclidean motion group. These include the evaluation of reachable positions and orientations at the distal end of a robot manipulator arm. A natural inverse problem arises when one wishes to design rather than to model manipulators. Namely, by considering a serial-chain robot arm as a concatenation of segments, we examine how statistics of known segments can be used to select, or design, the remainder of the structure so as to attain the desired statistical properties of the whole structure. This is then a deconvolution density estimation problem for the Euclidean motion group. We prove several results about the convergence of these deconvolution estimators to the true underlying density under certain smoothness assumptions. A practical implementation to the design of planar robot arms is demonstrated.

In this paper, adaptive control of kinematically redundant robot manipulators is considered. An end-effector tracking controller is designed and the manipulator's kinematic redundancy is utilized to integrate a general sub-task controller for self-motion control. The control objectives are achieved by designing a feedback linearizing controller that includes a least-squares estimation algorithm to compensate for the parametric uncertainties. Numerical simulation results are presented to show the validity of the proposed controller.

This paper analyzes the stabilization problem from the energy point of view. Perturbations are detected by the gyros and categorized according to the constraints on the zero-moment point, energy, and walking pattern. Ankle torque is exerted to extend the linear inverted pendulum mode (LIPM). Compensation movement is computed according to the analysis on the energy of LIPM and the influence of disturbance to the energy. The experimental results from both the simulation and the physical robot not only proved effective but also explain various human reactions to disturbance in locomotion.

The workspace of a parallel manipulator is usually smaller than the size of the robot itself. It is important to derive new structures that enjoy the advantages of parallel manipulators and also have a large workspace. In this paper we present two configurations of similar structures RRRS and RRSR with rotating links. The RRRS structure has a relatively large workspace—larger than the size of the robot itself—which is not common in parallel robots. The inverse and forward kinematics of the robots are presented. The workspaces of the robots are compared to similar and well-known structures, such as Eclipse, Alizade, Delta, and Hexa robots.

Today mosaics are used as decorative elements both indoors and outdoors. Mosaics can be obtained by combining small pieces of stone, glass, wood and glazed tiles together to make some picture and illustration. Mosaic-making processes are carried out manually for thousand years. Complexities involved in piece selection, surface detection and mould fitting have limited the automation of this process. In this work, software that makes the mosaic patterning process automatic is developed. For this purpose, software named SMT V2 has been developed to obtain different mosaic patterning. The software also tested with experimental applications by selective compliance articulated robot arm (SCARA) robot on real environment using different shaped marble pieces and successful results are obtained.

In this paper, we develop a model predictive control (MPC) scheme for robots to follow a virtual leader. The stability of this control scheme is guaranteed by adding a terminal state penalty to the cost function and a terminal state region to the optimization constraints. The terminal state region is found by analyzing the stability. Also a terminal state controller is defined for this control scheme. The terminal state controller is a virtual controller and is never used in the control process. Two virtual leader-following formation models are studied. Simulations on different formation patterns are provided to verify the proposed control strategy.

In structural design of current complex mechanisms or robots like parallel kinematic machines (PKMs), surgical robots, and reconfigurable robots, there commonly exist some functional modules called complex joints (CJs). Each of them, consisting of several simple pairs and essentially a mechanism, plays the same and more important roles as simple joints in kinematics and dynamics. However, as the primarily important aspect in mechanism analysis, the type and mobility of these CJs are far from familiarity. Therefore, this paper aims at addressing the type and mobility of CJs. For this purpose, the concept and classification of CJs are first discussed, an effective method to analyze the mobility characteristics of these CJs is then developed based on the equivalent screw system. The advantage of this method is that it reveals mobility characteristics by using equivalent transformations of kinematic pair screw (KP-screw) and constraint screw (C-screw) systems. With this method, the mobility characteristics of some concrete CJs are obtained correspondingly.

Screw theory has demonstrated its wide applications in robot kinematics and statics. We aim to propose an intuitive geometrical approach to obtain the reciprocal screws for a given screw system. Compared with the traditional Plücker coordinate method, the new approach is free from algebraic manipulation and can be used to obtain the reciprocal screws just by inspecting the structure of manipulator. The approach is based on three observations that describe the geometrical relation for zero pitch screw and infinite pitch screw. Based on the observations, the reciprocal screw systems of several common kinematic elements are analyzed, including usual kinematic pairs and chains. We also demonstrate usefulness of the geometrical approach by a variety of applications in mobility analysis, Jacobian formulation, and singularity analysis for parallel manipulator. This new approach can facilitate the parallel manipulator design process and provide sufficient insights for existing manipulators.

A robot has been developed to maintain boiler water-cooling tubes. This robot has a double tracked moving mechanism, an ash cleaning device, a slag purging device, a tubes' thickness measurement device, a marking device, and a control system. This robot can climb up and down the tube wall. A method for the robot to complete many tasks for boiler maintenance in one process cycle is presented. The mechanism of the robot is described. Especially, a kind of special magnetic block structure is designed to obtain strong adhering force using permanent magnets. Experiments in laboratory and real thermal power station have verified that the robot cannot only climb on the surface of the tube wall smoothly, but also take heavy payloads for boiler maintenance operation.

We investigated and identified the conditions necessary for stable dynamic gait generation in biped robots from the mechanical energy balance point of view. The equilibrium point at impact in a dynamic gait is uniquely determined by two conditions; keeping the restored mechanical energy constant and settling the relative hip-joint angle to the desired value before impact. The generated gait then becomes asymptotically stable around the equilibrium point determined by these conditions. This is shown by a simple recurrence formula of the kinetic energy immediately before impact. We verified this stability theorem using numerical simulation of virtual passive dynamic walking. The results were compared with those for a rimless wheel and an inherent stability principle was derived. Finally, we derived a robust control law using a reference mechanical energy trajectory and demonstrated its effectiveness numerically.