K. J. Austin

Application of genetic algorithms to hypersonic flight control

The paper presents an application of genetic algorithms to the design of a longitudinal flight controller for a hypersonic accelerator vehicle which is to be used to launch small satellites. A feature of hypersonic air-breathing flight vehicles is the high level of engine integration with the airframe. As a result, maintenance of vehicle attitude is not simply an issue of stability, but also one of propulsive effectiveness, which itself varies with flight conditions and the vehicle attitude. There is therefore limited scope for departure from optimum operating conditions. This, together with the extreme flight conditions, performance uncertainty, and the inherent instability of the vehicle, contributes to a demanding control task. We examine the capacity of a genetic algorithm in designing a fuzzy logic controller for the task of closed loop flight control. With a fixed, preset control structure, the design task is to configure the control surface through selection of the rule consequents and input scaling. The genetic algorithm uses a collection of simulated flight response in its formulation of the objective function. This allows the generation of a controller design without linearization of the vehicle model and dynamics. Stability augmentation is shown through flight simulation at the low-speed end of the hypersonic trajectory and also at a higher flight speed.

How do ICP variants perform when used for scan matching terrain point clouds

Many variants of the Iterative Closest Point (ICP) algorithm have been proposed for registering point clouds. This paper explores the performance of 20,736 ICP variants applied to the registration of point clouds for the purpose of terrain mapping, using data obtained from a mobile platform. The methodology of the study has involved taking sequences of 100 consecutive scans at three distinct scenes along the route of a mining haul truck operating in a typical surface mining environment. The scan sequences were obtained at 20 Hz from a Velodyne HDL-64E mounted on the truck. The aim is to understand how well the ICP variants perform in consolidating these scans into sub-maps. Variants are compared against three metrics: accuracy, precision, and relative computational cost. The main finding of the paper is that none of the variants is simultaneously accurate, precise, and fast to compute, across all three scenes. The best performing variants employed strategies that filtered the data sets, used local surface geometry in the form normals, and used the distance between points in one point cloud to a corresponding surface from a reference point cloud as a measure of the fit between two point clouds. The significance of this work is that it: (i) provides guidance in the construction of ICP variants for terrain mapping; and (ii) identifies the significant limitations of existing ICP variants for this application.

Swing Trajectory Control for Large Excavators

There is a strong push within the mining sector to automate equipment such as large excavators. A challenging problem is the control of motion on high inertia degrees of freedom where the actuators are constrained in the power they can deliver to and extract from the system and the machine’s underlying control system sits between the automation system and the actuators. The swing motion of an electric mining shovel is a good example. This paper investigates the use of predictive models to achieve minimum time swing motions in order to address the question what level of performance is possible in terms of realizing minimum time motions and accurate positional control. Experiments are described that explore these questions. The work described is associated with a project to automate an electricmining shovel and whilst the control law discussed here is a much simplified form of that used in this work, the experimental study sheds considerable light on the problem.

Comparison of numerical and experimental drag measurement in hypervelocity flow

In planning interplanetary missions which involve an aerobraking manoeuvre, it is necessary to make accurate predictions of the aerodynamic drag acting on a vehicle during its descent. Of interest to the authors is the Nasa initiative for exploration of Mars and its atmosphere. The Mars Pathfinder is a probe that is expected to enter the Martian atmosphere at a relative velocity of approximately 7.6 kms -1 ;. The forebody of this vehicle is based on a 70° blunted cone and is typical of aerobraking designs. In this note, a comparison is made between experimental and numerical techniques for predicting drag in hypervelocity flow. Three different models were examined in this study: a 30° sharp cone; an Apollo heat shield; and a Viking heat shield. A relatively simple analytical result for the drag on a cone provides a convenient reference for both the experimental and numerical results. The two heat shields are typical of those used for interplanetary exploration, such as the Mars Pathfinder. Our aim is to give an example of how computational fluid dynamics can be used in conjunction with experiments to obtain information about the hypervelocity flow about re-entry vehicles.

Transmission friction in an electric mining shovel

Friction is a phenomenon present in all mechanical systems and is undesirable in estimation and control tasks. This paper explores the characteristics of friction in the transmission of an electric mining shovel, with a view to compensating for them. Experiments suggest that the transmission friction can be described by a Coulomb friction model with associated uncertainty on the friction force. The Kalman Inverse Filter is evaluated as a method for adaptively estimating the friction for motion of the bucket through free-space. The friction estimates are consistent with experimental results

Establishment of Test Conditions in the RHYFL-X Facility

While current high enthalpy facilities are capable of generating the enthalpies associated with very high speed atmospheric flight, they are incapable of generating the large total pressures needed for full flow similarity. For accurate aerodynamic, heating, and combustion testing in ground-based facilities, it is desirable to increase their total pressure simulation capabilities. This paper continues the simulation study for the proposed conversion of RHYFL, originally designed to be a shock tube, into an expansion tunnel. In particular, we focus on the design of a suitable nozzle shape and on the details of the nozzle starting processes when supplied with a short duration pulse of test gas. To check whether steady flow conditions can be established in scramjet models, a ducted flow experiment has been conducted in a large prototype facility, X3. Flow measurements of heat transfer and static pressure indicate that useful test flows can indeed be established.

Monte-Carlo Tree Search in Dragline Operation Planning

Draglines are one of the largest earthmoving machines in surface mining. They are employed to remove the overburden near the surface, giving access to a seam of the target mineral or coal underneath. Their excavation productivity is significantly affected by the high-level operation strategy of where to position the dragline, what to dig at each position and where to dump the removed material. This letter explores the potential of using the Monte-Carlo tree search (MCTS) algorithm for planning this operation strategy. To this end, we adapt the MCTS algorithm to compute the dragline positioning sequences when a fixed digging and dumping strategy is applied at each position. The performance of the adapted MCTS algorithm is compared to a previously developed A* search algorithm and a greedy algorithm. Simulation results show that the MCTS algorithm is able to find near-optimal positioning sequences using significantly less time and memory than the A* algorithm. The solutions from the MCTS algorithm also largely outperform the solutions from the greedy algorithm in all the test scenarios. These results suggest that the MCTS algorithm can be used to plan the operation strategy when decisions about the digging and dumping operations are also considered.

Three new Iterative Closest Point variant-methods that improve scan matching for surface mining terrain

The Iterative Closest Point (ICP) algorithm seeks to minimize the misalignment between two point cloud data sets. A limitation of many ICP algorithms is that they work well for some contexts, yet perform poorly in others. Previous work has suggested that the ability of ICP variants to find correspondence was hindered by the presence of geometric disorder in the scene. This paper introduces three new methods based on characterizing the geometric properties of a point using information of its nearest neighbours. Two methods are entropy based and quantify the geometric disorder (eigentropy) in order to improve the filtering of data and thereby remove points that are likely to provide spurious associations. The third method is a point matching method using normals to preferentially work with planar areas of a point cloud. A set of 73,728 ICP variants obtained by combination/permutation of 26 methods are evaluated. These variants were evaluated using a scan matching exercise requiring construction of terrain maps based on data from a mobile sensing platform in an open-cut mining environment. The proposed methods improve ICP performance, as measured by accuracy, precision, and computational efficiency. Notably, five ICP variants, each featuring the new methods of this paper, simultaneously met the solution requirements for three different terrain scenes. It is asserted that being able to characterize the geometric disorder in the point clouds improves the capability of ICP to establish associations between points. 73,728 ICP variants were evaluated in three different mining scenes.The eigentropy filter is introduced for quantifying the geometrical disorder in point clouds.Five variants satisfy all performance criteria.Three new methods for the ICP pipeline are presented.