Jay Katupitiya

Estimating contact and grasping uncertainties using Kalman filters in force controlled assembly

This paper presents a general approach for the identification of contact and grasping uncertainties, and the monitoring of contact situation changes in force controlled assembly operations. The identification problem is solved using virtual contact manipulators and Kalman filter techniques. Monitoring is solved by carrying out a statistical test on the sum of normalized and squared innovations of the Kalman filter, within a moving window, identification and monitoring are verified by experimental results. The paper explains how the error covariance matrix of the Kalman filter is interpreted to analyse the observability of the Kalman filters states. Preliminary simulation results are presented for an ad hoc active sensing strategy to achieve complete state observability.

Robust Adaptive Coordination Controller for a Spacecraft Equipped with a Robotic Manipulator

Free-flying robotic spacecraft play a significant role in the space industry today. Unlike ground-based robots, the manipulator motion in a space robot can cause undesirable disturbances to the spacecraft platform, causing its attitude to change, potentially disrupting communication and solar energy collection processes as a result. Thus, coordinated control of both the spacecraft attitude and the manipulator motion become essential for successful space operations. Though past research has developed dynamic models for spacecraft manipulators, the contribution of reaction wheels to the angular momentum of the entire system needs further consideration. This paper reformulates the dynamic equations of a free-flying space robot by taking the aspect of reaction wheels into account. A diagonalization method is introduced to decouple the highly nonlinear multi-input/multi-output system model. A novel adaptive variable structure control method is then applied to implement a robust coordination controller for the ...

Modelling of a 4WS4WD vehicle and its control for path tracking

This paper presents a dynamic model representing the rigid body motion of a four wheel steer and four wheel drive (4WS4WD) vehicle which also takes into account the rotational dynamics of the individual wheels as well as the tyre terrain interactions. It also presents a modified pursuit algorithm taking into account the kinematic constraints as well as a decentralized wheel propulsion control system. Simulation results are presented to show the model validation and the successful implementation of the path tracking algorithm.

Path planning and tracking of a 4WD4WS vehicle to be driven under force control

The aim of this paper is to investigate the possibility of developing a system where forces at the wheels of a ground vehicle are controlled to guide the vehicle along a predetermined path. Any deviations from the desired path is handled using Bézier curve segments that put the vehicle back on track. The Bézier curve segments determine the necessary forces at the wheels. The future aim is to implement independent force control at each of the drive/steering modules of the vehicle. The vehicle is considered to be a rigid body with known inertia and mass. The paper presents path generation using Bézier curves, the extraction of desired forces and the determination of desired steering and propulsion to achieve the desired forces at the wheels of the vehicle. The forces and the control inputs such as the steering angles are then applied to a four wheel driven and four wheel steered vehicle subjected to slip in an open loop setting. Note that this paper does not address the force control issues. Simulation data are generated to validate the proposed methodology.

Dynamic modelling and control of a vectored thrust aerial vehicle

This paper presents the modelling and control of a ducted fan UAV platform powered by three ducted fans. The primary aim is to develop a highly reliable navigation platform with spacious payload bay. The chosen configuration also provides forward flight with minimum pitch variation. A complete non-linear dynamic model is developed. Taking advantage of the redundancy of the control inputs, a novel control scheme is designed to control the position and attitude based on the structured H∞ synthesis. A way-point navigation level is utilised to carry out “turn-and-cruise” flight in order to follow the given way-points. The performance of the overall control law tested with the nonlinear vehicle model is presented to show the operational capabilities of the vehicle.

Online path tracking and motion optimization of a 4WS4WD vehicle

This paper presents a methodology to optimize the drive forces and steering angles for achieving accurate path tracking by a four wheel steer and four wheel drive (4WS4WD) vehicle. The 7-order Bézier curves are applied in path planning and online tracking, and used to obtain the kinematic and dynamic profiles for the guidance of vehicle. The 8-input dynamic model is developed and applied in an objective function, which is optimized by PSO algorithm to obtain the optimal driving and steering motions for tracking the planned path precisely. The slip angles and lateral forces are also considered in the dynamic model, and the vehicle motion optimization can be implemented in real time. Note that this paper does not address the implementation of the force and steering controllers at the wheels. Simulation results are provided to validate the proposed method.

Derivation of error models incorporating slippage for a tracked vehicle coupled to a steerable trailer

This paper first presents a complete kinematic model and then presents kinematic model based error and offset models for tracked vehicles towing wheeled trailers or implements. In contrast to wheeled vehicles, the tracked vehicles have two tracks on either side of the vehicle. Steering is almost always through skid steer. They are fast becoming popular in agricultural fields due to excellent traction to weight ratio. As agricultural operations are progressing towards autonomous operations, it is important to develop models that can be used in developing controllers. In this paper kinematic models are used to develop two types of models one called the error model and the other called the offset model. They are developed for a system comprising a tracked vehicle towing a steerable wheeled trailer. The error model uses the errors in x, y and θ as the state variables of the model. The offset model uses the path offset and the alignment offset as state variables. Simulation results are presented to validate the error and offset models.

A gain adaptive Variable Structure Control methodology for second order nonlinear dynamic systems

This paper presents a Variable Structure Control (VSC) law with a gain adaptation mechanism for second order nonlinear dynamic systems which contain parametric uncertainties and external disturbances. The proposed method exhibit improved settling time compared to Sliding Mode Control (SMC), and has the ability to converge to a given set point switching only once. The controller achieves this by adapting its gain in real-time to match the effects of changing external disturbances and parametric uncertainties. The acceleration of the system is always directed towards the origin of error phase plane, and the trajectory of the error dynamics follow a parabola like path during control, improving the settling time as a result. Simulations and experiments are carried out on an inverted pendulum system to prove the performance and the practical applicability of the proposed method.

Force control for path following of a 4WS4WD vehicle by the integration of PSO and SMC

ABSTRACT The aim of this paper is to present a novel control method for a four-wheel steer and four-wheel drive (4WS4WD) vehicle. The novelty is in the integration of sliding mode control (SMC) and particle swarm optimization (PSO) that is proposed to solve the control problem caused by the nonlinear, highly coupled and over-actuated characteristics of the four-wheel steer and four-wheel drive (4WS4WD) vehicle. The validity of the control method is evaluated by two criterions, namely path following performance assessed by the vehicles position errors with respect to the reference path, and motion quality reflected by the smoothness of vehicles velocities and accelerations. In vehicle modelling, a kinematic model and a dynamic model considering all slip forces are proposed for the controller design. Simulation results are provided to demonstrate the applicability of the proposed methodology and its robustness.

An Adaptive Variable Structure Control methodology for attitude and position control of a Vectored Thrust Aerial Vehicle

This paper presents an Adaptive Variable Structure Control (AVSC) methodology to autonomously control an UAV platform known as the Vectored Thrust Aerial Vehicle (VTAV). The proposed control algorithm eliminates chattering present in Sliding Mode Control (SMC), and improves the settling time of the VTAV system by driving the error states in a parabolic trajectory towards the origin of error space instead of constraining the error states to a predefined sliding manifold. The proposed methodology is also robust to external disturbances and parametric uncertainties. Simulation results are presented to demonstrate the successful operation of the designed control algorithm on the VTAV.