No abstract

This paper discusses a class of mission planning problems that generalizes the standard Multiple Depot Vehicle Routing Problem with Time Windows (MDVRPTW) to incorporate complicated technical constraints. These constraints are specified via Metric Temporal Logic (MTL). A tree search algorithm is provided to solve that novel MDVRPTW with MTL specifications (MDVRPMTL) to optimality. In this work, we use tree search algorithm to seek the optimal flyable trajectories for teams of UAVs starting from different depots to complete missions with complicated time and technical requirements. Then a Stochastic Dynamic Programming (SDP) based algorithm is proposed to dynamically tune the UAV teams in order to secure them against adversarial actions. Examples for practical mission planning problems, in which MTL is used as a high level language to specify complex mission tasks, are presented and discussed in the paper.

For many decades, the Vehicle Routing Problem (VRP) and its different variants have been studied and found applications in the real world. This paper briefly surveys VRP instances with applications to multi-objective Unmanned Aerial Vehicle (UAV) operations. Focusing on multi-objective multi-UAV mission planning problems, we try to take advantage of the literature in the VRP and its variants. We show that each military multi-UAV mission has its corresponding VRP variant. We present a novel algorithm that relies on an enhanced tree search algorithm to solve complex multi-UAV mission planning problems with complex constraints. In simulation, we introduce examples for practical problem sizes in military UAV applications.

No abstract

We address a dynamic configuration strategy for teams of Unmanned Air Vehicles (UAVs). A team is a collection of UAVs which may evolve through different organizations, called configurations. The team configuration may change with time to adapt to environmental changes, uncertainty, and adversarial actions. Uncertainty comes from the stochastic nature of the environment and from incomplete knowledge of adversary behaviors. To each configuration, there corresponds a set of different properties for the UAVs in the team. The design for the configuration control problem involves a distributed hierarchical control architecture where the properties of the system can be formally analyzed. We do this in the framework of dynamic networks of hybrid automata. We present results from simulation to demonstrate different scenarios for adversarial response.

No abstract

Abstract-The purpose of this paper is to demonstrate that having two classifiers, a trichotomous classifier (true, false, or unknown) with workload-independent performance that turns over the data classified as unknown to a binary classifier (true or false) with workload-dependent performance, gives superior classification performance (lower probability of misclassification) compared to a single dichotomous classifier. We relate the classifier's performance to the inherent difficulty of the classification task at hand (classifiability), and compare the performance of different classifiers.

Major abrupt-onset cataclysmic events such as earthquakes, storms, floods, etc., typically damage infrastructure, cause injury, trap numerous individuals, and result in a massive death toll. A prompt life-saving response is required to rescue those who are marooned or trapped under debris. The difference between life and death can be a matter of how fast search and rescue attempts are carried out. On the other hand, these life-saving search and rescue operations are faced with real-time dynamic changes in the disaster site, in addition to possible communication network failure. This paper proposes a novel vision-based robot platooning algorithm that is capable of maneuvering teams of search and rescue robots in a dynamic disaster site, under the worst-case scenario of no available communication network. The algorithm was tested to drive teams of Pioneer-P3Dx and Jackal robots in five real different challenging disaster sites. The proposed algorithm showed enough robustness in all experiments to adapt to the dynamic environmental changes and drove the platoon to the desired destinations even when the team leader was lost.