Operation Of Multiple Vehicles Research Articles

Multi-start team orienteering problem for UAS mission re-planning with data-efficient deep reinforcement learning

In this paper, we study the Multi-Start Team Orienteering Problem (MSTOP), a mission re-planning problem where vehicles are initially located away from the depot and have different amounts of fuel. We consider/assume the goal of multiple vehicles is to travel to maximize the sum of collected profits under resource (e.g., time, fuel) consumption constraints. Such re-planning problems occur in a wide range of intelligent UAS applications where changes in the mission environment force the operation of multiple vehicles to change from the original plan. To solve this problem with deep reinforcement learning (RL), we develop a policy network with self-attention on each partial tour and encoder-decoder attention between the partial tour and the remaining nodes. We propose a modified REINFORCE algorithm where the greedy rollout baseline is replaced by a local mini-batch baseline based on multiple, possibly non-duplicate sample rollouts. By drawing multiple samples per training instance, we can learn faster and obtain a stable policy gradient estimator with significantly fewer instances. The proposed training algorithm outperforms the conventional greedy rollout baseline, even when combined with the maximum entropy objective. The efficiency of our method is further demonstrated in two classical problems – the Traveling Salesman Problem (TSP) and the Capacitated Vehicle Routing Problem (CVRP). The experimental results show that our method enables models to develop more effective heuristics and performs competitively with the state-of-the-art deep reinforcement learning methods.

The future of autonomous underwater vehicle control

The future of autonomous underwater vehicle control

Optimal Motion Planning for Differentially Flat Systems Using Bernstein Approximation

This letter presents a computational framework to efficiently generate feasible and optimal trajectories for differentially flat autonomous vehicle systems. We formulate the optimal motion planning problem as a continuous-time optimal control problem, and approximate it by a discrete-time formulation using Bernstein polynomials. These polynomials allow for efficient computation of various constraints along the entire trajectory, and are particularly convenient for generating trajectories for safe operation of multiple vehicles in complex environments. The advantages of the proposed method are investigated through theoretical analysis and numerical examples.

Operating Multiple Semi-Autonomous Robots: Monitoring, Responding, Detecting

The concept of one operator controlling multiple unmanned ground vehicle (UGV) robots is examined in a simulation of a multiple UGV search task. Semi-autonomous robotic UGVs were simulated searching for radioactive targets while operators performed various sub-tasks including monitoring status, responding to prompted target decision events, and detecting contextual errors. Participants executed a series of scenarios using 1, 2, or 4 UGVs simultaneously. The detection of contextual errors is particularly difficult when multiple vehicles are being supported. Specific interface features designed to assist with multiple vehicle operations are discussed including a quick video playback (QVP) function.

Strategies for simultaneous multiple autonomous underwater vehicle operation and control

Under sampling of the coastal oceans remains a persistent problem for standard oceano-graphic measurement practice wherein an instrument package is tethered to a research vessel. The overhead costs associated with operating a large research vessel impose a strict minimum on the cost of data collected. Owing to the overheads, significant improvements in sampling technology on the tethered platform can only produce modest gains in the cost effectiveness. In contrast, untethered vehicles if operated simultaneously have the potential to increase cost effectiveness significantly by distributing the overhead costs over several sampling platforms. Furthermore, synoptic and pseudosynoptic data can be collected with multiple autonomous underwater vehicles (AUVs), thereby providing the type of information critical to dynamic process modeling unattainable with non-synoptic data. While the goal of simultaneous multiple-vehicle operation has been espoused over the last few years, AUV technology and practice have until recently been too immature to realize that potential. Recently, Florida Atlantic University (FAU) has developed a new series of modular AUVs with the express purpose of supporting multiple sensors and multiple-vehicle operation. This series of vehicle is called the Ocean Explorer of which three have been produced so far. This paper will explore some of the associated navigation, tracking, control and deployment problems associated with multiple-vehicle operation in coastal applications. In addition, the characteristics of the component level intelligent distributed control system, integrated data logger and vehicle control system will be discussed. In particular this paper will discuss how FAU has applied the concepts of elastic constraint propagation and the symmetric fuzzy decision-making model to AUV control systems. Some results of early experiments in synoptic data collection with a conductivity, temperature and depth (CTD) sensor using multiple AUVs for the determination of horizontal structure will be described.

Vehicle Management System for Operating Multiple Vehicles

Multiple vehicle operation (e. g. cooperative object search) using autonomous underwater vehicles is an expected paradigm for deep ocean survey and exploration.In this paper, the system which consists of homogeneous vehicles is introduced to make a robust multiple vehicle system. The vehicle control architecture implemented in each member of the vehicles includes two decision making modules : one refers the individual environmental information gotten by its own sensors, and the other shares information received by other vehicles. By separating the decision maker into these two modules, the mission operator can easily write programs for each vehicle's behavior to generate an appropriate total vehicles' behavior. The performance of the vehicle's controlled by the proposed system is demonstrated on a multi-vehicle simulator, and the total behavior shows that the vehicles generate a formation which has not been explicitly given.