The objective of this study is to compare the abilities of logistic, auto-logistic and artificial neural network (ANN) models for quantifying the relationships between land uses and their drivers. In addition, the application of the results obtained by the three techniques is tested in a dynamic land-use change model (CLUE-s) for the Paochiao watershed region in Taiwan. Relative operating characteristic curves (ROCs), kappa statistics, multiple resolution validation and landscape metrics were used to assess the ability of the three techniques in estimating the relationship between driving factors and land use and its subsequent application in land-use change models. The validation results illustrate that for this case study ANNs constitute a powerful alternative for the use of logistic regression in empirical modeling of spatial land-use change processes. ANNs provide in this case a better fit between driving factors and land-use pattern. In addition, auto-logistic regression performs better than logistic regression and nearly as well as ANNs. Auto-logistic regression and ANNs are considered especially useful when the performance of more conventional models is not satisfactory or the underlying data relationships are unknown. The results indicate that an evaluation of alternative techniques to specify relationships between driving factors and land use can improve the performance of land-use change models.

The Muskingum model is the most widely used method for flood routing in hydrologic engineering. However, the application of the model still suffers from a lack of an efficient method for parameter estimation. Particle swarm optimization (PSO) is applied to the parameter estimation for the nonlinear Muskingum model. PSO does not need any initial guess of each parameter and thus avoids the subjective estimation usually found in traditional estimation methods and reduces the likelihood of finding a local optimum of the parameter values. Simulation results indicate that the proposed scheme can improve the accuracy of the Muskingum model for flood routing. A case study is presented to demonstrate that the proposed scheme is an alternative way to estimate the parameters of the Muskingum model.

In the near future, multi-sensor fusion will be the core component to navigate the autonomous driving platforms. This paper proposes new strategies to cope with the integration of an inertial navigation system (INS), a global navigation satellite system (GNSS), and light detection and ranging (LIDAR) to achieve simultaneous localization and mapping (INS/GNSS/LiDAR SLAM) especially in GNSS challenging environments where GNSS signals are blocked or contaminated with reflected signals. The proposed strategies implement a high level of integration with various information received from multiple sensors to collectively compensate for the specific drawbacks of those sensors included in the integrated system. The first strategy is to solve the divergence and drift problems of SLAM using the initial pose information from INS and the proposed refreshing process using an INS/GNSS integrated system. In addition, an updated mechanization is designed to qualify those received measurements based on cross validation of separate types of data. This mechanization is to ensure all measurements are reliable for the Extended Kalman Filter (EKF) update process. Moreover, the SLAM-derived information plays a major role to recognize the vehicle movement which assists the system to accurately apply those appropriate vehicle motion constraint models. The preliminary results presented in this study illustrate that proposed algorithm performs superior than the traditional INS/GNSS integration scheme and provides absolute navigation accuracy of 2 meters and 0.6% of distance traveled in GNSS-denied as well as 1.2 meters in GNSS-hostile environments, respectively.