AI Bibliography
AI Bibliography |
|
 |
 |
 |
|
Lee, C. Y. H. (2021). Deployment planning for multiple marsupial robots. Oregon State. |
| Resource type: Journal Article BibTeX citation key: Lee2021a View all bibliographic details  |
Categories: Artificial Intelligence, Complexity Science, Computer Science, Data Sciences, Engineering, General, Geopolitical, Mathematics, Military Science Subcategories: Analytics, Autonomous systems, Deep learning, Drones, Edge AI, Internet of things, Machine learning, Military research, Mosaic warfare, Networked forces, Robotics, United States Creators: Lee Publisher: Collection: Oregon State |
| Attachments |
| Abstract |
|
Multi-robot systems are versatile and extremely capable of exploration tasks in complex environments. Increasingly sophisticated planners, which incorporate new features of a multi-robot system, are necessary for the operation of the systems. Marsupial robots are multi-robot systems consisting of a carrier robot (e.g., a ground vehicle), which is highly capable and has a long mission duration, and at least one passenger robot (e.g., a short-duration aerial vehicle) transported by the carrier. These heterogeneous robot systems are more flexible than traditional homogeneous multi-robot systems in handling the constraints of a complex environment. However, the physical coupling of a carrier robot and passenger robots requires planners that are capable of accounting for passenger robot deployment planning in exploration. In this thesis, we present two algorithms for deploying passenger robots from marsupial robot systems, culminating into a generalized deployment framework. First, we address the single marsupial robot case: deploying multiple passenger robots from a single carrier robot. We optimize the performance of passenger robot deployment by using an algorithm that reasons over uncertainty by exploiting information about the prior probability distribution of features of interest in the environment. Our algorithm is formulated as a solution to a sequential stochastic assignment problem (SSAP). The key feature of the algorithm is a recurrence relationship that defines a set of observation thresholds that are used to decide when to deploy passenger robots. Our algorithm computes the optimal policy in O(NR) time, where N is the number of deployment decision points and R is the number of passenger robots to be deployed. We conducted drone deployment exploration experiments on real-world data from the DARPA Subterranean challenge to test the SSAP algorithm. Our results show that our deployment algorithm outperforms other competing algorithms, such as the classic secretary approach and baseline partitioning methods, and is comparable to an offline oracle algorithm. Secondly, we expand to the multiple marsupial robot case, deploying multiple passenger robots from multiple carrier robots. While the single marsupial robot deployment algorithm is an optimal, polynomial-time solution, the multiple-carrier robot deployment problem is fundamentally harder as it requires addressing conflicts and dependencies between deployments of multiple passenger robots. We propose a centralized heuristic search algorithm for the multiple-carrier robot deployment problem that combines Monte Carlo tree search with a dynamic programming-based solution to the Sequential Stochastic Assignment Problem as a rollout action-selection policy. Our results with both procedurally-generated data and data drawn from the DARPA Subterranean Challenge Urban Circuit show the viability of our approach and substantial exploration performance improvements over alternative algorithms.
|