Fusion of linear and nonlinear functional model for cadastral network adjustment / Nazurah Syahzanani Amiruddin

The eCadaster system's development aims to modernize the current implementation by digitalizing the land record, aligning with advancements in information technology. Subsequently, the Coordinate Cadastral System (CCS) has been introduced to serve the purpose of enhancing the efficiency of the cadastral surveying procedure. Nonetheless, in accordance with DSMM Circular No. 1/2010, the NDCDB has not completely preserved the expected positional accuracy. To systematically detect outliers in cadastral records, it is essential to have well-distributed constraints that are connected to existing cadastral boundary marks. However, the traverse method can lead to error propagation and involves significant costs, time, and labor. Instead of relying on traditional traversing, this study aims to quantify the reliability of GNSS baseline vectors in constraining the cadastral network adjustment. Due to the significant discrepancy in functional models, this study begins with the identification of an algorithm to harmonize the linear (GNSS vectors) and non-linear (traverse) models. Later, with the intention to numerically verify the reliability of the hybrid solution, experiments have been performed by employing multi-range configurations, which are: i) 0.5km; ii) 1.0km; iii) 1.5km; and iv) 2.0km. To scrutinize the solution, additional experiments have been carried out by taking into account the variation of measurement time: i)10 minute; ii) 15 minutes; iii) 20 minutes; and iv) 30 minutes. With the vector displacement of less than 0.050m, it is proven that the hybrid algorithm is effective and the solution manage to preserve the positional accuracy. Consequently, the findings have demonstrated the reliability of a hybrid algorithm as an alternative for cadastral network adjustment. With less augmentation by error propagation and simple in implementation, the hybrid algorithm is expected to realize the solution of enhancing the positional accuracy of NDCDB.