# Browsing by Subject "Autonomous agents"

Now showing 1 - 4 of 4

- Results Per Page
1 5 10 20 40 60 80 100

- Sort Options
Ascending Descending

Item Action selection and coordination of autonomous agents for UAV surveillance(2011-12) Han, David Ching-Wey; Barber, K. Suzanne; Arapostathis, Aristotle; Aziz, Adnan; Lifschitz, Vladimir; Stone, PeterShow more Agents, by definition, (1) are situated in an environment upon which their actions affect changes and (2) have some level of autonomy from the control of humans or other agents. Being situated requires that the agent have a mechanism for sensing the environment as well as actuators for changing the environment. Autonomy implies that each agent has the freedom to make their own decisions. Rational agents are those agents that decide to execute actions that are in their “best interests” according to their desires, using a model of those desires on which they make those decisions. Action selection is complicated due to uncertainty when operating in a dynamic environment or where other actors (agents) can also influence the environment. This dissertation presents an action selection framework and algorithms that are (1) rational with respect to multiple desires and (2) responsive with respect to changing desires. Agents can use the concept of commitments, and the subsequent communication of those commitments, to coordinate their actions and reduce their uncertainty. Coordination is layered on top of this framework by describing and analyzing how commitments affect the agents’ desires in their action selection models. This research uses the domain of UAV surveillance to experimentally explore the balance between under-commitment and over-commitment. Where previous approaches concentrate on the semantics of commitment, this research concentrates on the pragmatics of commitment, describing how to use utility calculations to enable an agent to decide when making a commitment is in its best interests.Show more Item Collision avoidance techniques and optimal synthesis for motion planning applications(2019-05-09) Marchidan, Andrei; Bakolas, Efstathios; Akella, Maruthi; Humphreys, Todd; Sentis, Luis; Longoria, RaulShow more This dissertation focuses on the problem of motion planning for autonomous agents that are required to perform fast and reactive maneuvers. In realistic situations, this problem needs to be solved in real-time for environments that are both dynamic and partially known. The success of the provided motion plans also relies on the agent’s ability to accurately perform the prescribed maneuvers and, as such, consideration of the input constraints is often times necessary. The problem can be posed in two different ways: as a controllability problem, where trajectory generation is only concerned with satisfying the given boundary conditions, system constraints (dynamic and input constraints) and state constraints (forbidden areas in the state space); or as an optimal control problem, where the trajectory is also required to optimize some performance measure. The main contributions of this dissertation are two-fold. First, a new numerical technique is proposed for solving time-optimal control problems for an agent moving in a spatiotemporal drift field. The solution technique computes the minimum time function and the corresponding time-optimal feedback control law, while using an extremal front expansion procedure to filter out sub-optimal solutions. This methodology can be applied for a rich class of time-optimal control problems where the control input structure is determined by a parameter family of differential equations. To demonstrate its applicability, the numerical technique is implemented for the Zermelo navigation problem on a sphere and for the steering problem of a self-propelled particle in a flow field. Next, in the second part of this dissertation, the controllability problem in the presence of obstacles can be solved using local reactive collision avoidance vector fields. The proposed approach uses the concept of local parametrized guidance vector fields that are generated directly from the agent model and encode collision avoidance behaviors. Their generation relies on a decomposition of agent kinematics and on a proximity-based velocity modulation determined by specific eigenvalue functions. Further exploiting the modulation properties arising from the nature of these eigenvalue functions, curvature constraints can be guaranteed. Closed-form steering laws are determined in accordance with the computed collision avoidance vector fields and can provide the necessary avoidance maneuvers to guarantee problem feasibility. Throughout this dissertation, examples and simulation results in different types of environments are presented and discussed. In the final part of this dissertation, the motion planning problem is tackled for more complex environments. The two proposed methodologies for optimal control and for collision avoidance are combined to yield a hybrid controller that generates near-optimal feasible plans in the presence of multiple static and moving obstacles and of spatiotemporal drift fields.Show more Item Evolving scout agents for military simulations(2015-05) Boyles, Brian David; Miikkulainen, Risto; Ballard, DanaShow more Simulations play an increasingly significant role in training and preparing the military, particularly in environments with constrained budgets. Unfortunately, in most cases a small number of people must control a large number of simulated vehicles and soldiers. This often leads to micromanagement of computer-controlled forces in order to get them to exhibit the human-like characteristics of an enemy force. This thesis uses Neuroevolution of Augmenting Topologies (NEAT) to train neural networks to perform the role of scouts which analyze the terrain and decide where to place themselves to best observe the enemy forces. The main attribute that the scout agents consider is a vapor flow rate from the enemy starting location to their intended objective, which according to previous studies indicates likely chokepoints along the enemy route. This thesis experiments with different configurations of sensors and fitness functions in order to maximize how much of the enemy team is spotted over the course of the scenario. The results show that these agents perform better than randomly placed scouts and better than scouts deployed using heuristics in many situations, although not consistently so. Evolutionary optimization of scout agents using vapor flow is thus a promising approach for developing autonomous scout agents in military simulations.Show more Item Visibility optimization for autonomous exploration and surveillance-evasion games(2020-05-08) Ly, Long Louis; Tsai, Yen-Hsi R.; Ghattas, Omar; Topcu, Ufuk; Vouga, Paul E; Ward, RachelShow more This dissertation considers problems involving line-of-sight visibility. In the exploration problem, the agent must efficiently map out a previously unknown environment, using as few sensor measurements as possible. For the surveillance-evasion game, the agent must always maintain visibility of a moving adversary. We first derive algorithms from a theoretical perspective. Although the resulting solutions are optimal, they are expensive to compute. We propose efficient approximations to the optimal solutions. At the expense of optimality, these approximations provide reasonable solutions that enable near real-time performance. We leverage state-of-the-art machine learning techniques to scale to scenarios that were not previously feasible. Level set functions allow for efficient computation of visibility and are a natural representation for input to convolutional neural networks. Lastly, we consider the notion of the visibility of point clouds along rays. Using nearest neighbors along a set of randomly generated rays, we compute a signature tensor which encodes geometric and statistical information about the point cloud. The signature, in combination with a convolutional neural network, enables efficient and robust classification of point clouds. We present promising results in 2D and 3D.Show more