Biologically inspired joint control for a humanoid robot

The GuRoo is a 1.2 m tall, 23 degree of freedom humanoid constructed at the University of Queensland for research into humanoid robotics. The key challenge being addressed by the GuRoo project is the development of appropriate learning strategies for control and coordination of the robot's many joints. The development of learning strategies is seen as a way to side-step the inherent intricacy of modeling a multi-DOF biped robot. This paper outlines the approach taken to generate an appropriate control scheme for the joints of the GuRoo. The paper demonstrates the determination of local feedback control parameters using a genetic algorithm. The feedback loop is then augmented by a predictive modulator that learns a form of feed-forward control to overcome the irregular loads experienced at each joint during the gait cycle. The predictive modulator is based on the CMAC architecture. Results from tests on the GuRoo platform show that both systems provide improvements in stability and tracking of joint control.

Enhancing opportunities to feed forward ; developing problem solving skills using web based scenarios

Universities often struggle to satisfy students’ need for feedback. This is an area where student satisfaction with courses of study can be low. Yet it is clear that one of the properties of good teaching is giving the highest quality feedback on student work. The term ‘feedback’ though is most commonly associated with summative assessment given by a teacher after work is completed. The student can often be a passive participant in the process. This paper looks at the implementation of a web based interactive scenario completed by students prior to summative assessment. It requires students to participate actively to develop and improve their legal problem solving skills. Traditional delivery of legal education focuses on print and an instructor who conveys the meaning of the written word to students. Today, mixed modes of teaching are often preferred and they can provide enhanced opportunities for feeding forward with greater emphasis on what students do. Web based activities allow for flexible delivery; they are accessible off campus, at a time that suits the student and may be completed by students at their own pace. This paper reports on an online interactive activity which provides valuable formative feedback necessary to allow for successful completion of a final problem solving assignment. It focuses on how the online activity feeds forward and contributes to the development of legal problem solving skills. Introduction to Law is a unit designed and introduced for completion by undergraduate students from faculties other than law but is focused most particularly on students enrolled in the Bachelor of Entertainment Industries degree, a joint initiative of the faculties of Creative Industries, Business and Law at the Queensland University of Technology in Australia. The final (and major) assessment for the unit is an assignment requiring students to explain the legal consequences of particular scenarios. A number of cost effective web based interactive scenarios have been developed to support the unit’s classroom activities. The tool commences with instruction on problem solving method. Students then view the stimulus which is a narrative produced in the form of a music video clip. A series of questions are posed which guide students through the process and they can compare their responses with sample answers provided. The activity clarifies the problem solving method and expectations for the summative assessment and allows students to practise the skill. The paper reports on the approach to teaching and learning taken in the unit including the design process and implementation of the activity. It includes an evaluation of the activity with respect to its effectiveness as a tool to feed forward and reflects on the implications for the teaching of law in higher education.

A Flight control scheme to improve position tracking performance of rotary-wing UASs

This paper outlines an innovative and feasible flight control scheme for a rotary-wing unmanned aerial system (RUAS) with guaranteed safety and reliable flight quality in a gusty environment. The proposed control methodology aims to increase gust-attenuation capability of a RUAS to ensure improved flight performance when strong gusts occur. Based on the design of an effective estimator, an altitude controller is firstly constructed to synchronously compensate for fluctuations of the main rotor thrust which might lead to crashes in a gusty environment. Afterwards, a nonlinear state feedback controller is proposed to stabilize horizontal positions of the RUAS with gust-attenuation property. Performance of the proposed control framework is evaluated using parameters of a Vario XLC helicopter and high-fidelity simulations show that the proposed controllers can effectively reduce side-effect of gusts and demonstrate performance improvement when compared with the proportional-integral-derivative (PID) controllers.

"Ruby's Music Festival" : developing problem solving skills using online scenarios and creating opportunities to feed forward

One of the characteristics of good teaching is giving the highest quality feedback on student work but the term “feedback” is most commonly associated with summative assessment given by a teacher after work is completed. The student can often be a passive participant in the process. This article looks at the implementation of web based scenarios completed by students prior to summative assessment with the objective of improving legal problem solving skills. It examines the design process and the implementation of the problem solving activity and the approach to teaching and learning taken in the new law unit of which it is part. We argue that such activities are effective tools to feed forward and reflect on the implications for the effective teaching of law in higher education.

Image-based visual servoing for the super-orbital re-entry of Hayabusa spacecraft

This paper presents an image-based visual servoing system that was used to track the atmospheric Earth re-entry of Hayabusa. The primary aim of this ground based tracking platform was to record the emission spectrum radiating from the superheated gas of the shock layer and the surface of the heat shield during re-entry. To the author's knowledge, this is the first time that a visual servoing system has successfully tracked a super-orbital re-entry of a spacecraft and recorded its pectral signature. Furthermore, we improved the system by including a simplified dynamic model for feed-forward control and demonstrate improved tracking performance on the International Space Station (ISS). We present comparisons between simulation and experimental results on different target trajectories including tracking results from Hayabusa and ISS. The required performance for tracking both spacecraft is demanding when combined with a narrow field of view (FOV). We also briefly discuss the preliminary results obtained from the spectroscopy of the Hayabusa's heat shield during re-entry.

Flight validation of a feedforward gust-attenuation controller for an autonomous helicopter

This paper presents a practical scheme to control heave motion for hover and automatic landing of a Rotary-wing Unmanned Aerial Vehicle (RUAV) in the presence of strong horizontal gusts. A heave motion model is constructed for the purpose of capturing dynamic variations of thrust due to horizontal gusts. Through construction of an effective gust estimator, a feedback-feedforward controller is developed which uses available measurements from onboard sensors. The proposed controller dynamically and synchronously compensates for aerodynamic variations of heave motion, enhancing disturbance-attenuation capability of the RUAV. Simulation results justify the reliability and efficiency of the suggested gust estimator. Moreover, flight tests conducted on our Eagle helicopter verify suitability of the proposed control strategy for small RUAVs operating in a gusty environment.

Design of a gust-attenuation controller for landing operations of Unmanned Autonomous Helicopters

This paper presents an innovative and practical approach to controlling heave motion in the presence of acute stochastic atmospheric disturbances during landing operations of an Unmanned Autonomous Helicopter (UAH). A heave motion model of an UAH is constructed for the purpose of capturing dynamic variations of thrust due to horizontal wind gusts. Additionally, through construction of an effective observer to estimate magnitudes of random gusts, a promising and feasible feedback-feedforward PD controller is developed, based on available measurements from onboard equipment. The controller dynamically and synchronously compensates for aerodynamic variations of heave motion resulting from gust influence, to increase the disturbance-attenuation ability of the UAH in a windy environment. Simulation results justify the reliability and efficiency of the suggested gust observer to estimate gust levels when applied to the heave motion model of a small unmanned helicopter, and verify suitability of the recommended control strategy to realistic environmental conditions.

Fuzzy associative memory for humanoid robot joint control

Traditional approaches to joint control required accurate modelling of the system dynamic of the plant in question. Fuzzy Associative Memory (FAM) control schemes allow adequate control without a model of the system to be controlled. This paper presents a FAM based joint controller implemented on a humanoid robot. An empirically tuned PI velocity control loop is augmented with this feed forward FAM, with considerable reduction in joint position error achieved online and with minimal additional computational overhead.

Designing a new robust on-line secondary path modeling technique for feedforward active noise control systems

Several approaches have been introduced in the literature for active noise control (ANC) systems. Since the filtered-x least-mean-square (FxLMS) algorithm appears to be the best choice as a controller filter, researchers tend to improve performance of ANC systems by enhancing and modifying this algorithm. This paper proposes a new version of the FxLMS algorithm, as a first novelty. In many ANC applications, an on-line secondary path modeling method using white noise as a training signal is required to ensure convergence of the system. As a second novelty, this paper proposes a new approach for on-line secondary path modeling on the basis of a new variable-step-size (VSS) LMS algorithm in feed forward ANC systems. The proposed algorithm is designed so that the noise injection is stopped at the optimum point when the modeling accuracy is sufficient. In this approach, a sudden change in the secondary path during operation makes the algorithm reactivate injection of the white noise to re-adjust the secondary path estimate. Comparative simulation results shown in this paper indicate the effectiveness of the proposed approach in reducing both narrow-band and broad-band noise. In addition, the proposed ANC system is robust against sudden changes of the secondary path model.

Intelligent prognostics of machinery health utilising suspended condition monitoring data

The ability to forecast machinery failure is vital to reducing maintenance costs, operation downtime and safety hazards. Recent advances in condition monitoring technologies have given rise to a number of prognostic models for forecasting machinery health based on condition data. Although these models have aided the advancement of the discipline, they have made only a limited contribution to developing an effective machinery health prognostic system. The literature review indicates that there is not yet a prognostic model that directly models and fully utilises suspended condition histories (which are very common in practice since organisations rarely allow their assets to run to failure); that effectively integrates population characteristics into prognostics for longer-range prediction in a probabilistic sense; which deduces the non-linear relationship between measured condition data and actual asset health; and which involves minimal assumptions and requirements. This work presents a novel approach to addressing the above-mentioned challenges. The proposed model consists of a feed-forward neural network, the training targets of which are asset survival probabilities estimated using a variation of the Kaplan-Meier estimator and a degradation-based failure probability density estimator. The adapted Kaplan-Meier estimator is able to model the actual survival status of individual failed units and estimate the survival probability of individual suspended units. The degradation-based failure probability density estimator, on the other hand, extracts population characteristics and computes conditional reliability from available condition histories instead of from reliability data. The estimated survival probability and the relevant condition histories are respectively presented as “training target” and “training input” to the neural network. The trained network is capable of estimating the future survival curve of a unit when a series of condition indices are inputted. Although the concept proposed may be applied to the prognosis of various machine components, rolling element bearings were chosen as the research object because rolling element bearing failure is one of the foremost causes of machinery breakdowns. Computer simulated and industry case study data were used to compare the prognostic performance of the proposed model and four control models, namely: two feed-forward neural networks with the same training function and structure as the proposed model, but neglected suspended histories; a time series prediction recurrent neural network; and a traditional Weibull distribution model. The results support the assertion that the proposed model performs better than the other four models and that it produces adaptive prediction outputs with useful representation of survival probabilities. This work presents a compelling concept for non-parametric data-driven prognosis, and for utilising available asset condition information more fully and accurately. It demonstrates that machinery health can indeed be forecasted. The proposed prognostic technique, together with ongoing advances in sensors and data-fusion techniques, and increasingly comprehensive databases of asset condition data, holds the promise for increased asset availability, maintenance cost effectiveness, operational safety and – ultimately – organisation competitiveness.

Submarine automatic control

This paper investigates the automatic atti- tude and depth control of a torpedo shaped submarine. Both experimental results and dynamic simulations are used to tune feed- back control loops in order to obtain stable control of yaw, pitch and roll of the craft.

Robust control algorithm for grid-interfaced three-phase inverters with high-bandwidth LCL filter

The paper introduces the design of robust current and voltage control algorithms for a grid-connected three-phase inverter which is interfaced to the grid through a high-bandwidth three-phase LCL filter. The algorithms are based on the state feedback control which have been designed in a systematic approach and improved by using oversampling to deal with the issues arising due to the high-bandwidth filter. An adaptive loop delay compensation method has also been adopted to minimize the adverse effects of loop delay in digital controller and to increase the robustness of the control algorithm in the presence of parameter variations. Simulation results are presented to validate the effectiveness of the proposed algorithm.

Prediction of activity type in preschool children using machine learning techniques

Objectives Recent research has shown that machine learning techniques can accurately predict activity classes from accelerometer data in adolescents and adults. The purpose of this study is to develop and test machine learning models for predicting activity type in preschool-aged children. Design Participants completed 12 standardised activity trials (TV, reading, tablet game, quiet play, art, treasure hunt, cleaning up, active game, obstacle course, bicycle riding) over two laboratory visits. Methods Eleven children aged 3–6 years (mean age = 4.8 ± 0.87; 55% girls) completed the activity trials while wearing an ActiGraph GT3X+ accelerometer on the right hip. Activities were categorised into five activity classes: sedentary activities, light activities, moderate to vigorous activities, walking, and running. A standard feed-forward Artificial Neural Network and a Deep Learning Ensemble Network were trained on features in the accelerometer data used in previous investigations (10th, 25th, 50th, 75th and 90th percentiles and the lag-one autocorrelation). Results Overall recognition accuracy for the standard feed forward Artificial Neural Network was 69.7%. Recognition accuracy for sedentary activities, light activities and games, moderate-to-vigorous activities, walking, and running was 82%, 79%, 64%, 36% and 46%, respectively. In comparison, overall recognition accuracy for the Deep Learning Ensemble Network was 82.6%. For sedentary activities, light activities and games, moderate-to-vigorous activities, walking, and running recognition accuracy was 84%, 91%, 79%, 73% and 73%, respectively. Conclusions Ensemble machine learning approaches such as Deep Learning Ensemble Network can accurately predict activity type from accelerometer data in preschool children.

Ongoing Adaptation as a Feature of Complexity : Further Thoughts and Possible Ideas for Pedagogy in Physical Activity

This chapter takes as its central premise the human capacity to adapt to changing environments. It is an idea that is central to complexity theory but receives only modest attention in relation to learning. To do this we will draw from a range of fields and then consider some recent research in motor control that may extend the discussion in ways not yet considered, but that will build on advances already made within pedagogy and motor control synergies. Recent work in motor control indicates that humans have far greater capacity to adapt to the ‘product space’ than was previously thought, mainly through fast heuristics and on-line corrections. These are changes that can be made in real (movement) time and are facilitated by what are referred to as ‘feed-forward’ mechanisms that take advantage of ultra-fast ways of recognizing the likely outcomes of our movements and using this as a source of feedback. We conclude by discussing some possible ideas for pedagogy within the sport and physical activity domains, the implications of which would require a rethink on how motor skill learning opportunities might best be facilitated.