Search

This thesis contributes to non-linear continuous-discrete Kalman filtering of multiplex systems through the development of two main ideas, namely, integration of the unscented transforms with linearly implicit methods and incorporation of simulation errors in the state estimation problem. The newly developed techniques are then applied to the technically relevant problem of state estimation on the main components of a utility boiler. State estimators in industrial systems are used as soft-sensors in monitoring and control applications as the most cost effective and practical alternative to telemetering all variables of interest. One such example is in utility boilers where reliable and real-time data characterising its behaviour is used to detect faults and optimise performance. With respect to the state-of-the-art, state estimators display limitations in real-time applications to large-scale systems. This motivates theoretical developments in state estimation as a first part in this thesis. These developments are aimed at producing more practical and efficient algorithms in non-linear continuous discrete Kalman filtering for stiff large-scale industrial systems. This is achieved using two novel ideas. The first is to exploit the similarities between the extended and unscented Kalman filter in order to estimate the Jacobian required for linearly implicit schemes, thereby tightly coupling state propagation and continuous-time simulation. The second is to account for numerical integration error by appending a stochastic local error model to the system's stochastic differential equation. This allows for coarser integration time steps in systems that are otherwise only suited to relatively small step sizes, making the filter more computationally efficient without lowering its potential to construct accurate estimates. The second part of this thesis uses these algorithms to demonstrate the feasibility of on-line state estimation on the main components of a once-through utility power boiler that require in excess of a hundred state variables to capture its behaviour with adequate fidelity. Two separate models of the boiler are developed, a MATLAB® and a Flownex® model, comprising the economiser, evaporators, reheaters, superheaters and furnace. The mathematical MATLAB® model is better suited to real-time execution and is used in the filter. The more sophisticated model is based on a commercial thermal-hydraulic simulation environment, Flownex® , and is used to validate the mathematical modelling philosophies and construct filter observation data. After validating the performance of the filter against ground-truth data provided by the Flownex® model, the filter is demonstrated on historical plant data to illustrate its utility.

Over the past five decades, Autonomous Mobile Robots (AMRs) have been an active research field. Maps of high accuracy are required for AMRs to operate successfully. In addition to this, AMRs needs to localise themselves reliably relative to the map. Simultaneous Localisation and Mapping (SLAM) address the problem of both map building and robot localisation. When exploring large areas, Multi-Robot SLAM (MRSLAM) has the potential to be far more efficient and robust, while sharing the computational burden across robots. However, MRSLAM encounters issues such as difficulty in map fusion of multi-resolution maps, and unknown relative positions of the robots. This thesis describes a distributed multi-resolution map merging algorithm for MRSLAM. HectorSLAM, which is one of many single robot SLAM implementations, has demonstrated exceptional results and was selected as the basis for the MRSLAM implementation in this project. We consider the environment to be three-dimensional with the maps being constrained to a two-dimensional plane. Each robot is equipped with a laser range sensor for perception and has no information regarding the relative positioning of the other robots. The experiments were conducted both in simulation and a real-world environment. Up-to three robots were placed in the same environment with Hector-SLAM running, the local maps and localisation were then sent to a central node, which attempted to find map overlaps and merge the resulting maps. When evaluating the success of the map merging algorithm, the quality of the map from each robot was interrogated. Experiments conducted on up to three AMRs show the effectiveness of the proposed algorithms in an indoor environment.

The gravitation towards mobility-as-a service in railway transportation system can be achieved at low cost and effort using shared railway network. However, the problem with shared networks is the presence of the level crossings where railway and road traffic intersects. Thus, long waiting time is expected at the level crossings due to the increase in traffic volume and heterogeneity. Furthermore, safety and capacity can be severely compromised by long level crossing closing time. The emphasis of this study is to optimise the rail-road level crossing closing time in order to achieve improved safety and capacity in a heterogeneous railway network. It is imperative to note that rail-road level crossing system assumes the socio-technical and safety critical duality which often impedes improvement efforts. Therefore, thorough understanding of the factors with highest influence on the level crossing closing time is required. Henceforth, data analysis has been conducted on eight active rail-road level crossings found on the southern corridor of the Western Cape metro rail. The spatial, temporal and behavioural analysis was conducted to extract features with influence on the level crossing closing time. Convex optimisation with the objective to minimise the level crossing closing time is formulated taking into account identified features. Moreover, the objective function is constrained by the train's traction characteristics along the constituent segments of the rail-road level crossing, speed restriction and headway time. The results show that developed solution guarantees at most 53.2% and 62.46% reduction in the level crossing closing time for the zero and nonzero dwell time, respectively. Moreover, the correctness of the presented solution has been validated based on the time lost at the level crossing and railway traffic capacity consumption. Thus, presented solution has been proven to achieve at most 50% recovery of the time lost per train trip and at least 15% improvement in capacity under normal conditions. Additionally, 27% capacity improvement is achievable at peak times and can increase depending on the severity of the headway constraints. However, convex optimisation of the level crossing closing time still fall short in level crossing with nonzero dwell time due to the approximation of dwell time based on the anticipated rather than actual value.

Main steam temperatures play a significant role in large coal fired power plant operation. Ideally, main steam temperatures should be accurately controlled to protect the thick wall components against long term overheating and thermal stress while meeting the design conditions at the steam turbine inlet. Although high steam temperatures are beneficial for thermal efficiency, it accelerates creep damage in high temperature components which is detrimental to the life of components. Alternatively, low steam temperatures increase the moisture content at the last stage blades of the turbine, causing the blades to deteriorate and fail. Control of the outlet steam temperature according to design conditions at variable loads is maintained via a balance between heat input (flue gas temperature and mass flow rate), evaporator outlet steam mass flow and spray water. The present control philosophy accuracy of main steam temperatures at an Eskom coal fired power plant was evaluated and compared to the latest technology and control strategies. Improving and optimizing steam temperature controls ensures design efficiency while maintaining long term plant health. The level of spatial discretization applied in simplifying the real boiler for modelling purposes was approached at a relatively high level. The intention was to model normal operating conditions and certain transients such as variable heat input and load changes to see its effect on steam temperatures and to be able to evaluate the performance of different temperature control techniques. The main outcome of this project was to design a robust control system for a dynamic model of the boiler using sets of low order linear models to account for uncertainty. The main concepts, models and theories used in the development of this dissertation include: 1) A detailed thermo-fluid model developed using Flownex to have high fidelity models of the process under varying operating conditions. This model was used to test and evaluate the robust controller design. 2) System Identification in Matlab to construct mathematical models of dynamic systems from measured inputoutput data and identify linear continuous time transfer functions under all operating conditions [1]. 3) Quantitative Feedback Theory (QFT) to design controllers for an attemperator control system at various onload operating conditions. This design was used understand the engineering requirements and seeks to design fixed gain controllers that will give desired performance under all operating conditions. 4) The design of a valve position controller to increase the heat uptake in a convective pass, thereby improving efficiency: Excessive attemperation in the superheater passes is generally associated with high flue gas temperatures which decrease thermal efficiency. Therefore, robust control of the attemperation system leads to an increase in heat uptake between the flue gas and steam in the boiler, resulting in a reduction in the flue gas temperature leaving the boiler, thus improving efficiency. The robust QFT controllers were set up using the valve position control technique and were used to confirm the improvement of control performance. The theories mentioned above were used to understand the control performance under varying plant conditions using a standard cascaded arrangement. It incorporated robust control design and engineering requirements such as bandwidth, plant life, spray water and thermodynamic efficiency. The control effort allocated to each superheaterattemperator subsystem in the convective pass was designed as a multi-loop problem.

The prevalence of electrical transmission networks has led to an increase in productivity and prosperity. In 2014, estimates showed that the global electric power transmission network consisted of 5.5 million circuit kilometres (Ckm) of high-voltage transmission lines with a combined capacity of 17 million mega-volt ampere. The vastness of the global transmission grid presents a significant problem for infrastructure maintenance. The high maintenance costs, coupled with challenging terrain, provide an opportunity for autonomous inspection robots. The Brachiating Power Line Inspection Robot (BPLIR) with wheels [73] is a transmission line inspection robot. The BPLIR is the focus of this research and this dissertation tackles the problem of state estimation, adaptive trajectory generation and robust control for the BPLIR. A kinematics-based Kalman Filter state estimator was designed and implemented to determine the full system state. Instrumentation used for measurement consisted of 2 Inertial Measurement Units (IMUs). The advantages of utilising IMUs is that they are less susceptible to drift, have no moving parts and are not prone to misalignment errors. The use of IMU's in the design meant that absolute angles (link angles measured with respect to earth) could be estimated, enabling the BPLIR to navigate inclined slopes. Quantitative Feedback Control theory was employed to address the issue of parameter uncertainty during operation. The operating environment of the BPLIR requires it to be robust to environmental factors such as wind disturbance and uncertainty in joint friction over time. The resulting robust control system was able to compensate for uncertain system parameters and reject disturbances in simulation. An online trajectory generator (OTG), inspired by Raibert-style reverse-time symmetry[10], fed into the control system to drive the end effector to the power line by employing brachiation. The OTG produced two trajectories; one of which was reverse time symmetrical and; another which minimised the perpendicular distance between the end gripper and the power line. Linear interpolation between the two trajectories ensured a smooth bump-less trajectory for the BPLIR to follow.

Following a paper published by E. Boje[1], this thesis discusses the design and off-line testing of different types of Kalman filters to estimate the attitude, position and velocity of a robotic platform moving along a power line. The nature of this problem limits the use of magnetometers. Magnetic field interference from the steel pylons and steel cored conductors will affect the local magnetic field. Moreover, high frequency signals from on-board power electronic drives and induced magnetic fields due to ferromagnetic components of the robot along with aliasing, quantization effects and a low signal to noise ratio make notch filtering at 50 Hz impractical. Thus, a GPS/IMU filter solution, which uses the power line curvature and horizontal direction in measurements, to constrain the robot to the line was designed. Different types of filters were implemented; The Extended Kalman filter (EKF), the Unscented Kalman filter (UKF) and the Error State Kalman filter (ErKF). Measurements were recorded and the filters were tested offline. While all the filters tracked properly, it was found that the EKF was better in computational speed completing an iteration in 87 µs, the ErKF was second best with an average time of 120 µs for one iteration and the UKF was last with an average time of 1040 µs for one iteration. Errors between the true state and estimated state for the simulation were quantified using root mean square values (RMS). The RMS values were almost the same for the EKF and ErKF with the error for the x position at 0.81 m and z position at 0.038 m. The UKF produced RMS errors of 0.79 m for x position and 0.11 m for z position. It can be seen that the UKF is slightly better for the x position but is much worse for the z position. Overall, the GPS measurement RMS values used were 4 m and 20 m for the horizontal and vertical positions respectively. Thus, the filters brought a big improvement. However, the recommended filter is the EKF as is produced comparable or better results as compared to other filters and expends the least computational effort. A state estimator was also developed for a J.Patel’s PLIR project [2], where a brachiating version of a power line robot was modeled. The brachiation mechanism was approximated to a double pendulum and kinematics based Kalman filter was designed. Simulations of EKF and UKF were made. The EKF is still recommended as its estimates are closer to the true values and its computation time is about five times faster.

The development of a control system to transport and assemble cargo using two helicopters is presented in this thesis. It is more economical to use multiple lower cost helicopters in a coordinated manner to carry cargo than to use a single high performance helicopter for the transportation task. The reason for the generally higher cost of hiring high performance helicopters, is because they are not required often, and so, remain idle for most of their lifetime. Thus, using less specialised, lower performing helicopters to share the load is cheaper. Beyond just sharing the load of the cargo, the objective in this investigation is to control the attitude such that precise placement of the cargo can be made. This objective cannot be achieved using a single helicopter, unless a sophisticated tethering mechanism is developed. The installation of wind-turbine blades, powerline towers and radio masts in remote locations, are examples of where the application of this technology may be useful. The investigation of this thesis is around modelling, estimation and control of the twinhelicopter slung load transportation system. The title reflects the investigation that was required to be done to determine whether a scheme could be realisable. To test the concept, an experimental platform was developed. A small, light-weight and high performance avionics system was designed and interfaced to the helicopters. The experimentation was done indoors, and hence, the flying volume was limited. For the purpose of feedback and analysis, a motion capture system was developed to track the position and attitude of the helicopters. A high-fidelity mathematical model of a small-scale helicopter was developed. Estimation algorithms were then developed to optimally fuse the data from the instrumentation designed. The data was then used in a system identification exercise to find the parameters that capture the dynamics of the helicopter. The full constrained model of the twin-helicopter slung load dynamics was then developed. The high-fidelity multivariable, interacting system was then linearised to generate a set of uncertain plants. Unexpected resonant modes were investigated using modal analysis to understand their source. Robust controllers were designed using Quantitative Feedback Theory (QFT) for the individual helicopter attitude and altitude loops. A solution was found for the twin-helicopter load transportation system by decoupling the plant with a static pre-compensator and then designing a decentralised QFT controller for the 6 × 6 plant. The effort of this thesis is towards the (practical) realisation of a twin-helicopter aerial crane capable of attitude control; the architecture for the industrialisation of the twin-helicopter load transportation system is proposed.

A new class of unmanned marine research vessels makes use of wave propulsion to minimise energy requirements during voyages. Existing models of these hybrid sea-surface and underwater craft have not considered if the platform’s interaction with the immediate surrounding sea could be incorporated to allow for more accurate navigation and path planning. To this end a detailed three-dimensional model of one such vessel, the Liquid Robotics’ Wave Glider, has been developed in this study. The multi-body system is described using DenavitHartenberg parametrisation and a Lagrangian approach is used to generate the equations of motion for the body. Physical dimensions are derived from platform measurements and from the product specification sheet, hydrodynamic factors are derived from a SolidWorks model of the system, and added mass components are determined from empirical data. Finally, the dynamic model is verified for a given sea state and multiple sea states are tested to investigate the effect on the model’s performance. The developed Wave Glider model is shown to have a realistic response when hydrodynamic factors, added mass and hydrodynamic damping forces, are included and to sea states in terms of the hydrostatic restorative response. The wave-driven propulsion provided by the hydrofoils is shown to have dependence on the sea state by running the model in an open-loop simulation. Following the model validation, a control system is developed for the Wave Glider model to allow yaw attitude control of the glider using the controllable glider rudder input. The control system is generated making use of quantitative feedback theory (QFT) methods to provide robust control for the under-actuated system. The control scheme is shown to provide suitable performance for sea states that result in variable glider velocities. The model’s performance, in terms of the average velocity, is shown to have dependence on the direction of the sea state by running the model in an open-loop simulation for multiple sea states with sinusoidal waves approaching the Wave Glider model from different directions.

This dissertation aims to apply non-zero attitude and position setpoint tracking to a quadrotor aircraft, achieved by solving the problem of a quadrotor’s inherent underactuation. The introduction of extra actuation aims to mechanically accommodate for stable tracking of non-zero state trajectories. The requirement of the project is to design, model, simulate and control a novel quadrotor platform which can articulate all six degrees of rotational and translational freedom (6-DOF) by redirecting and vectoring each propeller’s individually produced thrust. Considering the extended articulation, the proposal is to add an additional two axes (degrees) of actuation to each propeller on a traditional quadrotor frame. Each lift propeller can be independently pitched or rolled relative to the body frame. Such an adaptation, to what is an otherwise well understood aircraft, produces an over-actuated control problem. Being first and foremost a control engineering project, the focus of this work is plant model identification and control solution of the proposed aircraft design. A higher-level setpoint tracking control loop designs a generalized plant input (net forces and torques) to act on the vehicle. An allocation rule then distributes that virtual input in solving for explicit actuator servo positions and rotational propeller speeds. The dissertation is structured as follows: First a schedule of relevant existing works is reviewed in Ch:1 following an introduction to the project. Thereafter the prototype’s design is detailed in Ch:2, however only the final outcome of the design stage is presented. Following that, kinematics associated with generalized rigid body motion are derived in Ch:3 and subsequently expanded to incorporate any aerodynamic and multibody nonlinearities which may arise as a result of the aircraft’s configuration (changes). Higher-level state tracking control design is applied in Ch:4 whilst lower-level control allocation rules are then proposed in Ch:5. Next, a comprehensive simulation is constructed in Ch:6, based on the plant dynamics derived in order to test and compare the proposed controller techniques. Finally a conclusion on the design(s) proposed and results achieved is presented in Ch:7. Throughout the research, physical tests and simulations are used to corroborate proposed models or theorems. It was decided to omit flight tests of the platform due to time constraints, those aspects of the project remain open to further investigation. The subsequent embedded systems design stemming from the proposed control plant is outlined in the latter of Ch:2, Sec:2.4. Such implementations are not investigated here but design proposals are suggested. The primary outcome of the investigation is ascertaining the practicality and feasibility of such a design, most importantly whether or not the complexity of the mechanical design is an acceptable compromise for the additional degrees of control actuation introduced. Control derivations and the prototype design presented here are by no means optimal nor the most exhaustive solutions, focus is placed on the whole system and not just a single aspect of it.

Ocean observation is vital in understanding how the oceans contribute toward climate change and other effects. This is one of many undertakings requiring a persistent presence in the oceans. These maritime activities are mainly carried out on large research vessels chartered for weeks at a time, which can be extremely costly. In addition, the data obtained when using these vessels are only short snapshots of the continual processes that occur. Recently, there has been a drive toward using Unmanned Surface Vehicles (USVs) and Unmanned Underwater Vehicles (UUVs), which can be deployed at a fraction of the cost, and provide greatly improved spatio-temporal data. The wave glider (WG) is one such autonomous marine robot used for persistent ocean research and other maritime activities, and forms the focus of this study. The WG is a low power USV/UUV hybrid that harnesses wave energy for propulsion, and has a small solar- and battery-powered thruster, and a rudder for steering. Due to effects of waves, currents, and other disturbances, the platform tends to veer off its desired path. Additionally, local sea state information is not taken into consideration while manoeuvring, hence energy extraction from ocean waves is not optimal. More sophisticated navigation algorithms operating on a per-wave strategy may improve accuracy along a specified path and maximise the energy uptake from the waves. To realise these improvements requires prediction of local wave behaviour. If one can predict what the wave field will be a short time in the future, then possible control action can be taken to efficiently navigate in the environment. Inertial measurements and wave modelling have been used to improve localisation of the WG platform directly, and predict the platform’s velocity. However there is limited work in the context of WG navigation. Hence the problem this dissertation aims to solve is the estimation and subsequent prediction of local wave behaviour. This work proposes a novel approach to estimate the sea state and hence predict short-term, local wave behaviour from inertial measurements on a slow-moving marine platform such as the WG. A Kalman filtering strategy consisting of a phase-locked loop and filter based sea state estimator is used to generate local height and angle of arrival estimates. This method offers an improvement over existing Fast Fourier Transform methods as it does not require long time series data to produce results, and enables the prediction of wave behaviour a short time into the future. The ideas are tested in simulation by generating wind waves using ocean wave models such as the Pierson Moskowitz model, and dynamic a dynamic model of the WG platform. In addition, a small scale lab experiment is carried out to verify the performance of the sea-state estimator developed. Preliminary results obtained indicate that relative wave height can be estimated on-board a marine platform, using only inertial sensors.

The inspection of power lines and associated hardware is vital to ensuring the reliability of the transmission and distribution network. The repetitive nature of the inspection tasks present a unique opportunity for the introduction of robotic platforms, which offer the ability to perform more systematic and detailed inspection than traditional methods. This lends itself to improved asset management automation, cost-effectiveness and safety for the operating crew. This dissertation presents the development of a prototype industrial brachiating robot. The robot is mechanically simple and capable of dynamically negotiating obstacles by brachiating. This is an improvement over current robotic platforms, which employ slow, high power static schemes for obstacle negotiation. Mathematical models of the robot were derived to understand the underlying dynamics of the system. These models were then used in the generation of optimal trajectories, using nonlinear optimisation techniques, for brachiating past line hardware. A physical robot was designed and manufactured to validate the brachiation manoeuvre. The robot was designed following classic mechanical design principles, with emphasis on functional design and robustness. System identification was used to capture the plant uncertainty and a feedback controller was designed to track the reference trajectory allowing for energy optimal brachiation swings. Finally, the robot was tested, starting with sub-system testing and ending with testing of a brachiation manoeuvre proving the prospective viability of the robot in an industrial environment.

The cheetah (Acinonyx jubatus) is by far the most manoeuvrable and agile terrestrial animal. Little is known, in terms of biomechanics, about how it achieves these incredible feats of manoeuvrability. The transient motions of the cheetah all involve rapid flicking of its tail and flexing of its spine. The aim of the research was to develop tools (hardware and software) that can be used to gain a better understanding of the cheetah tail and spine by capturing its motion. A mechanical rig was used to simulate the tail and spine motion. This insight may inspire and aid in the design of bio-inspired robotic platforms. A previous assumption was that the tail is heavy and acts as a counter balance or rudder, yet this was never tested. Contrary to this assumption, necropsy results determined that the tail was in fact light with a relatively low inertia value. Fur from the tail was used in wind tunnel experiments to determine the drag coefficient of a cheetah tail. No researchers have actively sought to track the motion of a cheetah's spine and tail during rapid manoeuvres via placing multiple sensors on a cheetah. This requires the development of a 3D dynamic model of the spine and tail to accurately study the motion of the cheetah. A wireless sensor network was built and three different filters and state estimation algorithms were designed and validated with a mechanical rig and camera system. The sensor network consists of three sensors on the tail (base, middle and tip) as well as a hypothetical collar sensor (GPS and WiFi were not implemented).