Differential global positioning system

Abstract

Differential Global Positioning System (DGPS) is emerging as a crucial technology within the resource and communications sector, allowing for the development of autonomous control in hazardous environments. This thesis is concerned with the design, construction and evaluation of a low cost and high performance DGPS which can be embedded into robotic systems. The principle behind this thesis is to prove that a DGPS can be designed to meet the same performance of an industrial grade system, while using low end components. The thesis is composed of five major stages: research, design, construction, evaluation and documentation. This thesis is to provide documentation of the development and evaluation of the DGPS. The final product is a fully documented and highly functional DGPS system, for the Murdoch University School of Engineering & Energy to integrate into future projects. Standard GPS systems are based on a single Global Navigation Satellite System (GNSS) signal receiver to process position co-ordinates and relevant data. These GNSS signals containing pseudoranges, travel through the atmosphere and are susceptible to distortion, which is the predominate cause of error in GPS positioning. The atmospheric distortion and hence positioning error can be significantly reduced through the use of DGPS as it assumes that the signals have travelled the same atmospheric path and hence induced with the same distortion. By adopting this concept, highly accurate relative positioning between receivers can be produced through a series of calculations. DGPS can be extended further to produce corrected positioning based on fixed receiver locations which can compare the error of the fixed position to those received, and then apply the same error to other mobile receiver positions. Currently a large scale network of DGPS base stations exists and can be used for correcting commercial and industrial grade systems, however these can be expensive. The expense of these systems is largely dependent on the accuracy and functionality that the system can produce, and therefore highlights the need for this thesis as a means of a cost effective solution. The prototype produced in this thesis was the product of thorough engineering design and evaluation of the three major components; communications, hardware and software. Each of these components had sub-objectives allow the main objective of a low cost, high performance system to be achieved. Evaluation of proposed communication wireless transceivers was undertaken to find characteristics of reliability, suitable bandwidth and a sufficient operating distance. The wireless communications is used for the transmission of the base station positioning data to the mobile station for processing. Comparative testing methods of the APPCON APC200 found it to be a suitable alternative to the University proposed HopeRF HM-TR transceivers. The APPCON provides reliable data transmission, with error correction capabilities while exceeding an operational distances of 850 meters. Processing the incoming simultaneous stream of pseudoranges from both GPS receiver modules requires a minimum of two hardware universal asynchronous receiver/transmitter (UART) ports, while meeting the system memory and processing requirements. The initial open-source Arduino Duemilanove microcontroller chosen for the project was deemed to be unsuitable after evaluation of two software based UART. The Arduino Mega, having met these requirements, has been used as the prototype microcontroller as it provides four hardware UART ports and largely redundant memory and processing capabilities. Arduino is based on the open-source software environment called ‘wiring’ which is a derivative of the C++ software language. The strong online support community for the Arduino has developed an expansive range of open-source software libraries which can be adapted to user projects. Implementing these software libraries, developed as community based projects, means that advanced capabilities and extensively tested code is integrated into the system software code. The outcome of combining these components has completed the sub objectives of this project and has resulted in a high functioning, reliable, accurate and low cost system. Thorough evaluation of the DGPS prototype has yielded results that substantially exceed the manufacturer rated specifications and prove the benefits of DGPS relative positioning. The average accuracy of the system in an open environment has achieved 0.216 meter averaged distance and an operation range of 882 meters. Testing of the system has been undertaken in a number of environments to evaluate the accuracy and reliability of the system. The finalized hardware value of the prototype was $467.28, developing an accurate positioning system at a budget price.