Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 24 for “"partially observable Markov decision process (POMDP)"”.
-
Planning under uncertainty for dynamic collision avoidance
… We first formulate the problem within the Partially Observable Markov Decision Process (POMDP) framework, and use generic MDP/POMDP solvers offline to compute vertical-only avoidance strategies that optimize a cost function to balance flight-plan deviation with risk of collision. We then …
-
Multi-objective optimization of next-generation aircraft collision avoidance software
… shown that formulating collision avoidance as a partially-observable Markov decision process (POMDP) can dramatically increase system performance. However, the POMDP formulation relies on a number of design parameters modifying these parameters can dramatically alter system behavior. Prior work …
-
Modeling Human Planning in Maze Orienteering Problems
… build models of human planning given their past decisions. In this thesis project, I focused on modeling human planning in Maze Orienteering Problems (MOP), an optimization problem with the objective to maximize collected rewards within a time limit in a partially known maze. The project has two …
-
Probabilistic roadmaps in uncertain environments
… of some edges are unknown. This is modelled as a decision-theoretic planning problem described through a partially observable Markov Decision Process (POMDP). It is shown that the optimal policy can depend on accounting for the value of information from observations. The model scalability and the …
-
Reactive, Autonomous, Markovian Sensor Tasking in Communication Starved Environments
… for evaluating these techniques. A suboptimal partially observable Markov decision process (POMDP) was used as the simulation framework to test various reward functions and decision algorithms while enabling autonomous, reactive sensor tasking. The goal of this work was used the developed …
-
Relatively robust grasping
… while grasping, we model the problem as a partially observable Markov decision process (POMDP). We derive a closed-loop strategy that maintains a belief state (a probability distribution over world states), and select actions with a receding horizon using forward search through the belief …
-
Managing resources on a multi-modal sensing device for energy-aware state estimation
… to existing sensor scheduling methods for hidden Markov models. We extend these methods, and cast the problem as a standard partially observable Markov decision process (POMDP), for which numerous exact and approximate solutions are well known. We then demonstrate optimal sensing policies on a …
-
Path planning and control of flying robots with account of human’s safety perception
… using a virtual reality environment. A hidden Markov model (HMM) is considered for estimation of latent variables, as user’s attention, intention, and emotional state. Then, an optimal motion planner generates a trajectory, parameterized in Bernstein polynomials, which minimizes the cost …
-
Planning with imperfect information : interceptor assignment
… of RVs. This work formulates the problem as a partially observable Markov decision process (POMDP) in order to account for the uncertainty in information. We use a POMDP solution algorithm to find an optimal policy for assigning interceptors to RVs in a single wave. From there, three cases are …
-
Automated planning for hydrothermal vent prospecting using AUVs
… to a problem that is naturally formulated as a partially-observable Markov decision process (POMDP), but with a very large state space (of the order of 10\(^{123}\) states). This size of problem is intractable for current POMDP solvers, so instead heuristic solutions were sought. The problem is …
-
Deep Recurrent Q Networks for Dynamic Spectrum Access in Dynamic Heterogeneous Envirnments with Partial Observations
… approaches cannot solve the resulting online Partially-Observable Markov Decision Process (POMDP), Deep Recurrent Q-Networks (DRQN) have been proposed to determine the optimal channel access policy via online learning. The fundamental goal of this dissertation is to develop DRL-based solutions …
-
A POMDP framework for antenna selection and user scheduling in multi-user massive MIMO systems
We use a partially observable Markov decision process (POMDP) framework to design a resource allocation policy for downlink transmit beamforming at a multi-antenna BS that is equipped with a massive number of antennas and only a limited number of RF chains. Considering that channels evolve …
-
Partially Observable Markov Process Decision Modeling for the Optimal Maintenance of Oil and Gas Pipelines
Partially Observable Markov Decision Process (POMDP) frameworks are employed across various fields. This dissertation studies the applications of POMDP for maintaining oil and gas pipelines. Pipeline maintenance operations comprise several uncertain elements, especially when addressing …
-
Learning sparse representation for image signals
… successfully exploited in a variety of image processing applications, ranging from low level recovery to high level semantic inference. A good sparse representation is expected to have high fidelity to the observed image content and at the same time reveal the underlying structure and semantic …
-
ML-Based Optimization of Large-Scale Systems: Case Study in Smart Microgrids and 5G RAN
… such as face recognition and natural language processing, have demonstrated the great potential of ML techniques. Indeed, ML can significantly enhance the intelligence of many existing systems, including smart grid, wireless communications, mechanical engineering, and so on. For instance, …
-
Knowledge and Ignorance in Reinforcement Learning
… concerned with teaching agents to take optimal decisions to maximize their total utility in complicated environments. A Reinforcement Learning problem, generally described by the Markov Decision Process formalism, has several complex interacting components, unlike in other machine learning …
-
Adaptive robotic search and sampling of sparse natural phenomena
… of these approaches produce maps of easily observable and widely dispersed phenomena such a temperature, salinity or tree coverage. However underwater and planetary science can often involve phenomena that are ‘expensive’ to observe, discrete, and sparsely distributed. For example, coral …
-
Computational Techniques for Stochastic Reachability
… stochastic reachable sets for both perfectly and partially observable systems. We initially consider a linear system with additive Gaussian noise, and introduce two methods for computing stochastic reachable sets that do not require dynamic programming. The first method uses a particle …
-
Adaptive Robotic Search and Sampling of Sparse Natural Phenomena
… of these approaches produce maps of easily observable and widely dispersed phenomena such a temperature, salinity or tree coverage. However underwater and planetary science can often involve phenomena that are ‘expensive’ to observe, discrete, and sparsely distributed. For example, coral …
-
Managing heterogeneous resources for dynamic energy-efficient sensing
… for a system with a given set of sensing and processing resources. The utilization problem is mapped to a partially observable Markov decision process (POMDP) and the appropriate mapping is derived in order to leverage state-of-the-art POMDP numerical solvers to generate optimal …
Page 1 of 2