Gait Pattern Adaptation for an Active Lower-Limb Orthosis Based on Neural Networks

This work deals with neural network (NN)-based gait pattern adaptation algorithms for an active lower-limb orthosis. Stable trajectories with different walking speeds are generated during an optimization process considering the zero-moment point (ZMP) criterion and the inverse dynamic of the orthosis–patient model. Additionally, a set of NNs is used to decrease the time-consuming analytical computation of the model and ZMP. The first NN approximates the inverse dynamics including the ZMP computation, while the second NN works in the optimization procedure, giving an adapted desired trajectory according to orthosis–patient interaction. This trajectory adaptation is added directly to the trajectory generator, also reproduced by a set of NNs. With this strategy, it is possible to adapt the trajectory during the walking cycle in an on-line procedure, instead of changing the trajectory parameter after each step. The dynamic model of the actual exoskeleton, with interaction forces included, is used to generate simulation results. Also, an experimental test is performed with an active ankle–foot orthosis, where the dynamic variables of this joint are replaced in the simulator by actual values provided by the device. It is shown that the final adapted trajectory follows the patient intention of increasing the walking speed, so changing the gait pattern.     

Hardness prediction models based on HAZ simulation for in-service welded pipeline steels

Abstract

Welding onto pressurised pipelines that contain flammable fluid to facilitate repairs or branch connections, is a critical procedure with considerable risk to personnel and infrastructure. Limiting the heat input is obviously an important consideration to prevent 'burn-through', but the potential for rapid cooling of the weldment increases its susceptibility to hydrogen assisted cold cracking (HACC). Therefore, one of the most important factors for in-service welding procedure development relates to the increased risk of formation of hard, brittle microstructures in the grain coarsened heat affected zone (GCHAZ) of the weld, microstructures that increase the risk of HACC. The present work has been successful in utilising heat treatment simulations to derive two new hardness prediction models that more accurately predict hardness in the GCHAZ for typical in-service welding applications than a commercially adopted and widely used hardness prediction algorithm. Although it is acknowledged that further work is required to validate the models for a wider range of in-service welding conditions, the performance of the models demonstrates their potential for developing improved in-service welding procedures.     

Open-end human–robot interaction from the dynamical systems perspective: mutual adaptation and incremental learning

In this paper, we experimentally investigated the open-end interaction generated by the mutual adaptation between humans and robot. Its essential characteristic, incremental learning, is examined using the dynamical systems approach. Our research concentrated on the navigation system of a specially developed humanoid robot called Robovie and seven human subjects whose eyes were covered, making them dependent on the robot for directions. We used the usual feed-forward neural network (FFNN) without recursive connections and the recurrent neural network (RNN) for the robot control. Although the performances obtained with both the RNN and the FFNN improved in the early stages of learning, as the subject changed the operation by learning on its own, all performances gradually became unstable and failed. Next, we used a 'consolidation-learning algorithm' as a model of the hippocampus in the brain. In this method, the RNN was trained by both new data and the rehearsal outputs of the RNN not to damage the contents of current memory. The proposed method enabled the robot to improve performance even when learning continued for a long time (open-end). The dynamical systems analysis of RNNs supports these differences and also showed that the collaboration scheme was developed dynamically along with succeeding phase transitions.     

Neural network face detection

Abstract

Analysing a video sequence is a challenge, because faces are constantly in dynamic motion, presenting many different possible rotational and illumination conditions. While solutions to the task of face detection have been presented, the detection performances of many systems are heavily dependent upon a strictly constrained environment. This paper presents the results of an image-based neural network face detection system which seeks to address the problem of detecting faces under gross variations.     

A System for Robotic Heart Surgery that Learns to Tie Knots Using Recurrent Neural Networks

Tying suture knots is a time-consuming task performed frequently during minimally invasive surgery (MIS). Automating this task could greatly reduce total surgery time for patients. Current solutions to this problem replay manually programmed trajectories, but a more general and robust approach is to use supervised machine learning to smooth surgeon-given training trajectories and generalize from them. Since knot tying generally requires a controller with internal memory to distinguish between identical inputs that require different actions at different points along a trajectory, it would be impossible to teach the system using traditional feedforward neural nets or support vector machines. Instead we exploit more powerful, recurrent neural networks (RNNs) with adaptive internal states. Results obtained using long short-term memory RNNs trained by the recent Evolino algorithm show that this approach can significantly increase the efficiency of suture knot tying in MIS over preprogrammed control.     

Hardware implementation of neuromodulated neural network for a CPU-less autonomous mobile robot

In order to construct truly autonomous mobile robots, the concept of 'packaging' is indispensable; in packaging, all parts such as controllers, power systems and batteries should be embedded inside a finite physical space, i.e., a robot's body. Therefore, implementing a controller on hardware is one of the most promising ways, since this contributes to low power consumption, miniaturization, etc. Another crucial requirement in the field of autonomous mobile robots is robustness, i.e., autonomous mobile robots have to cope with their unpredictably changing environment in real-time. In this study, to meet these requirements, the concept of a dynamically rearrangeable electrical circuit (DREC) is proposed and we implement this onto field progammable gate arrays as physical electronic circuits by borrowing the idea from neuromodulation widely observed in biological nervous systems through the diffusion-reaction mechanism of neuromodulators. We developed the DREC for the peg-pushing task as a practical example. We confirmed that the physical DREC can successfully regulate the behavior according to the situation encountered by changing its properties in real-time.     

Predicting Object Dynamics From Visual Images Through Active Sensing Experiences

Prediction of dynamic features is an important task for determining the manipulation strategies of an object. This paper presents a technique for predicting dynamics of objects relative to the robot's motion from visual images. During the training phase, the authors use the recurrent neural network with parametric bias (RNNPB) to self-organize the dynamics of objects manipulated by the robot into the PB space. The acquired PB values, static images of objects and robot motor values are input into a hierarchical neural network to link the images to dynamic features (PB values). The neural network extracts prominent features that each induce object dynamics. For prediction of the motion sequence of an unknown object, the static image of the object and robot motor value are input into the neural network to calculate the PB values. By inputting the PB values into the closed loop RNNPB, the predicted movements of the object relative to the robot motion are calculated recursively. Experiments were conducted with the humanoid robot Robovie-IIs pushing objects at different heights. The results of the experiment predicting the dynamics of target objects proved that the technique is efficient for predicting the dynamics of the objects.     

Fuzzy self-adaptive radial basis function neural network-based control of a seven-link redundant industrial manipulator

This paper proposes a method for the identification of dynamics and control of a multi-link industrial robot manipulator using Runge-Kutta-Gill neural networks (RKGNNs). RKGNNs are used to identify an ordinary differential equation of the dynamics of the robot manipulator. A structured function neural network (NN) with sub-networks to represent the components of the dynamics is used in the RKGNNs. The sub-networks consist of shape adaptive radial basis function (RBF) NNs. An evolutionary algorithm is used to optimize the shape parameters and the weights of the RBFNNs. Due to the fact that the RKGNNs can accurately grasp the changing rates of the states, this method can effectively be used for long-term prediction of the states of the robot manipulator dynamics. Unlike in conventional methods, the proposed method can even be used without input torque information because a torque network is part of the functional network. This method can be proposed as an effective option for the dynamics identification of manipulators with high degrees-offreedom, as opposed to the derivation of dynamic equations and making additional hardware changes as in the case of statistical parameter identification such as linear least-squares method. Experiments were carried out using a seven-link industrial manipulator. The manipulator was controlled for a given trajectory, using adaptive fuzzy selection of nonlinear dynamic models identified previously. Promising experimental results are obtained to prove the ability of the proposed method in capturing nonlinear dynamics of a multi-link manipulator in an effective manner.     

A tree architecture with hierarchical data processing on a sensor-rich hexapod robot

This paper presents the hardware and software architecture of Golem, a hexapod robot designed as a flexible, scalable, general purpose development and experimenting tool targeted to academia, industry, and defense environments. The system is technologically innovative in its architecture, performance, size and integration, and is industrially promising in its filling the gap between low-performance commercial solutions and costly application-specific proprietary solutions.     

A Model-Based Performance Management Tool for ATM and Frame Relay Networks

In this paper we describe a network modelingapproach intended to assist in the performancemanagement, design, and optimization of broadbandtraffic networks. Switch and source models, as well asrouting optimization and decision support algorithmshave been integrated in a prototype software tool,called DATANMOT (Data Network Modeling and OptimizationTool). The switch models developed are based on standard Frame Relay and ATM switch implementations.Specifically, an analytical model of the Fujitsu FETEX150 ATM switch is described here in detail. Fluid-flowapproximation methods were used for performance evaluation, with computational complexity lowenough for near-real time applications. As a result,given the network configuration and input traffic, anevaluation of the quality of service can be derived and used in optimal routing, admission controland network planning. These techniques have beenincorporated in our modeling tool to demonstrate themodel-based approach to network management. In addition, all configuration, modeling, and managementfunctions of the software tool are supported by agraphical user interface, and a databasesystem.