Multi-Agent Network Control - A Brain Emotional Learning-Inspired Approach
Abstract
The research objective of this project is to employ computational models of emotional learning observed in the mammalian limbic system to develop novel and systematic methodologies for analysis, design, and implementation of autonomous multi-agent systems (MAS) operations. The motivation comes from the interdisciplinary and complex nature of the tasks encountered in the modern society, which demand the integration of multiple complementary agents capable of self-organization and coordinate themselves. This work builds on the PI s previous results demonstrating that the implementation of Brain Emotional Learning (BEL) is effective in stabilizing a single autonomous aerial agent, as well as a team of robots, both in simulations and real-time experiments. Coordination of MAS in real- time missions is challenging because the dynamics of the robotic agents, which could be aerial, ground, water vehicles, or even a combination of them, are usually not precisely known. Furthermore, MAS operations are often subjected to external disturbances and varying operational conditions. It is hypothesized that BEL strategies will provide MAS with learning capabilities, multi-objective properties, and low computational complexity. To achieve these goals, the proposed project will focus on the following fundamental research questions: 1. Adaptation of reward/sensory signals: BEL controllers must be provided with sensory signals and emotional cues, which should make sense with respect to the MAS states/objectives. It is hypothesized that, by adaptively choosing these functions by means of self- organizing methods and multi-objective goals (flocking, obstacle avoidance, energy savings) the mission will be optimally performed. 2. Implementation of BEL-inspired MAS control: Exploiting the small computational cost and adaptive capacity of the proposed BEL control, we will show the applicability of our approach in realistic MAS applications, e.g., cooperative load transportation. A network of aerial/ground robotic vehicles will be used in a laboratory for validation purposes. This project aims at an integrated design to achieve MAS controllers with efficiency, reliability and flexibility, with less complexity and lower computational costs. The integration of BEL in MAS control is still unexplored. Transformative project outcomes will advance knowledge in the MAS domain, establishing foundations for future exploration of complementary methodologies to improve this field. Integrated education/outreach activities will support the Texas A&M University - Corpus Christi (TAMU-CC) Unmanned Aircraft Systems Summer Institute (UASSI), an engineering event happening every year in July, and targeting local K-12 underrepresented students. Students from the Society of Hispanic Professional Engineers will be involved as mentors for working with UASSI attendants. UASSI activities will increase awareness and inform young students of the needs in the domain of complex networks, and will increase participation of underrepresented students in engineering and computing sciences. This research plan will support the recently created TAMU-CC Ph.D. program in Geospatial Computing Sciences and will advance the TAMUCC s effort to establish the nation s first Ph.D. program in Autonomous Vehicles (expected Fall 2018). Relevance to the Network Sciences Division: The development of distributed BEL feedback controllers for MAS subjected to uncertain dynamics and disturbances contributes to the field of intelligent networked systems, and directly addresses the Multi-Agent Network Control programmatic aims. BEL-inspired distributed control for MAS is a novel and multidisciplinary research area, which advances ties will be conducted in the Unmanned Systems Laboratory led by the PI, and working on collaborative projects with faculty in the departments of ENCS, Mathematics, and Psychology.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 14, 2019
- Source ID
- W911NF1810210
Entities
People
- Luis Rodolfo García Carrillo
Organizations
- Army Contracting Command
- Texas A&M University–Corpus Christi
- United States Army