Abstract
The integration of unmanned ground-based robotic systems into logistics and other automation-intensive sectors has notably expanded in recent years. Despite their increasing prevalence, the lack of a unified classification framework has complicated the systematic selection and deployment of these platforms. This research has been initiated to address this gap by developing the EPAM classification system, which incorporates Environment, Purpose, Autonomy, and Mobility as core multidimensional criteria. The purpose of this study is to develop a coherent and technically grounded taxonomy for Unmanned Ground Vehicles (UGVs), with a particular focus on logistics and multifunctional applications. The methodology has involved a comprehensive literature review, systematization of existing classification approaches, and a comparative analysis of their scope and limitations. The EPAM framework has been defined in terms of discrete sets of application environments (e.g., paved roads, dirt roads, rough terrains), intended functions (e.g., transport, patrol, combat, supply), autonomy levels (from remote-controlled to fully autonomous), and platform mobility types (wheeled, tracked, legged, hybrid). Key findings have demonstrated that current classification systems fail to standardize vital technical parameters such as drive configuration, modularity, and autonomy architecture. Therefore, A consistent taxonomy has been proposed, enhancing technical communication between developers and users and supporting more informed decision-making. The novelty of the approach lies in its formalized, multidimensional structure, which enables the coverage of real-world UGV application scenarios and environmental constraints, including transitional terrains, wet surfaces, and complex, obstacle-ridden areas. A preliminary classification of existing UGV platforms has been initiated to verify and validate the applicability of the EPAM model. This step will serve as empirical support for the robustness and scalability of the proposed system. The practical value of the study is reflected in its potential to foster interoperability standards, enhance procurement specifications, and expedite the development of adaptive UGV platforms for dynamic logistics operations. Future investigations are expected to focus on refining classification criteria based on operational feedback, integrating Artificial Intelligence (AI)–based perception technologies, and applying the EPAM system in simulation and field-testing environments. The proposed framework offers a foundation for further scientific exploration in ground robotics and operational planning under uncertainty.