Unmanned Aerial Vehicle (UAV) photogrammetry is increasingly used for local-scale mapping because it enables rapid generation of high-resolution orthophotos and elevation products. However, the positional and vertical quality of these products depends strongly on the georeferencing strategy, particularly the use of Ground Control Points (GCPs). This study evaluates the effect of GCPs on the accuracy of UAV-derived orthophoto, Digital Surface Model (DSM), and Digital Terrain Model (DTM) at the premises of the Land Management Training Center (LMTC), Dhulikhel, Nepal. The same UAV image block of 334 images was processed under two workflows: one using only onboard image geolocation and another using five surveyed GCPs. Eight independent checkpoints were used for accuracy assessment. Orthophoto planimetric accuracy was evaluated using checkpoint coordinates, while vertical agreement of DSM and DTM was assessed using surveyed ground elevations. In addition, pairwise checkpoint-distance analysis and supplementary object-based comparison were used to examine relative geometric differences in the orthophotos. The with-GCP orthophoto achieved a horizontal RMSEH of 0.063 m, whereas the without-GCP orthophoto showed an RMSEH of 3.507 m. Similarly, the with-GCP DSM and DTM achieved RMSEZ values of 0.131 m and 0.143 m, respectively, compared with 15.385 m and 15.347 m in the without-GCP workflow. The without-GCP orthophoto also exhibited systematic westward and northward displacement and minor scale-related geometric differences. The results demonstrate that GCP-based processing remains essential for reliable campus-scale UAV photogrammetry when the outputs are intended for measurement, terrain representation, planning, and other applications requiring dependable absolute accuracy.
B. Bisht, Nabraj Subedi, Ram Kumar Sapkota et al.· Journal of Land Management a...· 0 citations
Agentic Artificial Intelligence (AI) represents a paradigm shift from static, task-specific systems to autonomous, goal-directed agents capable of reasoning, planning, learning, and acting with minimal human oversight. This paper synthesizes perspectives from philosophy, cognitive science, and AI to define agency, outline its key properties, and situate it in relation to existing paradigms such as reinforcement learning, symbolic reasoning, Belief–Desire–Intention (BDI) architectures, and embodied cognition. We present a taxonomy of agentic systems along dimensions of autonomy, cognitive capability, modality, and environmental interaction, highlighting current capabilities and limitations, and we critically delimit where such a taxonomy is informative and where a functional, closed-loop analysis of agent behavior must take over. The enabling technologies, including large language models, memory architectures, planning frameworks, tool-use mechanisms, and multimodal embodiment, are reviewed alongside diverse application domains ranging from autonomous research and creative systems to web automation and human–AI collaboration. We analyze safety, alignment, evaluation challenges, and emergent risks, dedicate a section to trustworthiness, privacy, and sustainability by bridging from trustworthy machine learning, verified autonomy, and privacy-preserving learning, and compile a comparative review of prominent benchmarks for assessing agentic behavior. Finally, we outline future research directions, including compositional and modular architectures, cooperative AI, simulation-based safe exploration, data-efficient and developmentally inspired learning, hybrid symbolic–neural systems, and strategies for robust alignment. By providing a comprehensive framework and critical analysis, this work aims to guide the development of agentic AI systems that are not only capable but also safe, trustworthy, and aligned with human values.