Dr. David K.A. Mordecai Co-Advised an NYU CUSP Capstone on Autonomous Drone Swarms
Dr. David K.A. Mordecai and Professor Giuseppe Loianno, PhD, Professor of Engineering at NYU Tandon School of Engineering co-advised a team of graduate students in an NYU Center for Urban Science and Progress (CUSP) Capstone research project entitled Applying Multi-Agent RL to SLAM with Graph Pose for Sampled-Data MPC and CPN of Autonomous Drone Swarms, with the undernoted synopsis.
Synopsis: Developing methods that allow drones to autonomously navigate in different environments has been a topic of extensive research in recent years. One research topic of interest for interest in autonomous drone navigation is to explore the maneuverability and capability of drones to navigate inaccessible environments and situations that might be too risky for human access. Deploying a drone swarm to autonomously navigate a post-catastrophe scenario in order to optimally map the disaster zone, i.e., independently and efficiently identify and map the structural damage across a geographic site, has been a problem less explored. Detection and mapping changes across a post-catastrophe site enables a more robust estimation of structural damage. This project attempted to explore and simulate a reinforcement learning approach to enable drones to perform task assignment and scheduling in order to efficiently maximize coverage for identifying and mapping structural changes within the post-catastrophe environment. The primary objective of the simulation was to focus on the exploration of ad-hoc decentralized task assignment and scheduling by one or more drone(s) at the edge with minimal connectivity aside from local communication between nearest neighbors. Other workstreams in the project preliminarily explored satellite and aerial imagery, seismic structural damage equation models and generative adversarial networks (GANs) related to the 2010 Port-au-Prince Haiti earthquake site as a case study, and methods that might be utilized to identify structural changes from satellite images, using generative synthetic data and estimated fragility equations in order to address uncertainty and ambiguity in the detection of discrepancies in edges related to damage which could aid the drones in the uncertain areas.
Dr. Mordecai is Co-Managing Member of Numerati® Partners and advises research activities at RiskEcon® Lab @ Courant Institute of Mathematical Sciences NYU.
About Numerati® Partners LLC
Numerati® Partners LLC coordinates a data analytics and technology development ecosystem, with the mission of advancing and fostering the next generation of scalable data-intensive risk and liability management enterprises. The firm provides resources fundamental to advancing the development of nascent leading-edge inferential surveillance, monitoring, and predictive analytics technologies for deployment within the RiskTech domain: risk technologies associated with adaptive distributed, networked and embedded systems such as remote sensing, agent-oriented data analytics, computing and control systems. Numerati® Partners curates integrated RiskTech solutions as well as forensic and use-case applications in RiskTech sub-domains such as LitTech, RegTech, FinTech and InsurTech (litigation technology, regulation technology, financial technology and insurance technology).
About RiskEcon® Lab @ Courant Institute
The mission of RiskEcon® Lab for Decision Metrics @ Courant Institute of Mathematical Sciences NYU is the development of experimental testbeds and analytics that employ high-dimensional datasets from innovative sources by applying a range of computational and analytical methods to commercial and industrial sensor networks and edge-computing for distributed, embedded and autonomous systems, focusing primarily on research and development (R&D) of remote and compressed sensing, anomaly detection, forensic analytics and statistical process control. By employing applied computational statistics within the context of robust and scalable data analytic solutions, our goal is robust integration of machine learning with signal processing for measurement and control, in order to conduct research fundamental to large-scale, real-world questions in risk and liability management.
RiskEcon® Lab enables, facilitates and coordinates academic research focusing on these patterns and trends, through the development of commercially-viable, analytic applications employing computational statistical tools in conjunction with innovative and non-traditional data structures. In addition, the Lab’s activities involve the advancement of applied mathematical statistics and computational economics, through interdisciplinary post-doctoral, postgraduate, graduate research and education in data science and social computing in the public interest.
Synopsis: Developing methods that allow drones to autonomously navigate in different environments has been a topic of extensive research in recent years. One research topic of interest for interest in autonomous drone navigation is to explore the maneuverability and capability of drones to navigate inaccessible environments and situations that might be too risky for human access. Deploying a drone swarm to autonomously navigate a post-catastrophe scenario in order to optimally map the disaster zone, i.e., independently and efficiently identify and map the structural damage across a geographic site, has been a problem less explored. Detection and mapping changes across a post-catastrophe site enables a more robust estimation of structural damage. This project attempted to explore and simulate a reinforcement learning approach to enable drones to perform task assignment and scheduling in order to efficiently maximize coverage for identifying and mapping structural changes within the post-catastrophe environment. The primary objective of the simulation was to focus on the exploration of ad-hoc decentralized task assignment and scheduling by one or more drone(s) at the edge with minimal connectivity aside from local communication between nearest neighbors. Other workstreams in the project preliminarily explored satellite and aerial imagery, seismic structural damage equation models and generative adversarial networks (GANs) related to the 2010 Port-au-Prince Haiti earthquake site as a case study, and methods that might be utilized to identify structural changes from satellite images, using generative synthetic data and estimated fragility equations in order to address uncertainty and ambiguity in the detection of discrepancies in edges related to damage which could aid the drones in the uncertain areas.
Dr. Mordecai is Co-Managing Member of Numerati® Partners and advises research activities at RiskEcon® Lab @ Courant Institute of Mathematical Sciences NYU.

About Numerati® Partners LLC
Numerati® Partners LLC coordinates a data analytics and technology development ecosystem, with the mission of advancing and fostering the next generation of scalable data-intensive risk and liability management enterprises. The firm provides resources fundamental to advancing the development of nascent leading-edge inferential surveillance, monitoring, and predictive analytics technologies for deployment within the RiskTech domain: risk technologies associated with adaptive distributed, networked and embedded systems such as remote sensing, agent-oriented data analytics, computing and control systems. Numerati® Partners curates integrated RiskTech solutions as well as forensic and use-case applications in RiskTech sub-domains such as LitTech, RegTech, FinTech and InsurTech (litigation technology, regulation technology, financial technology and insurance technology).
About RiskEcon® Lab @ Courant Institute
The mission of RiskEcon® Lab for Decision Metrics @ Courant Institute of Mathematical Sciences NYU is the development of experimental testbeds and analytics that employ high-dimensional datasets from innovative sources by applying a range of computational and analytical methods to commercial and industrial sensor networks and edge-computing for distributed, embedded and autonomous systems, focusing primarily on research and development (R&D) of remote and compressed sensing, anomaly detection, forensic analytics and statistical process control. By employing applied computational statistics within the context of robust and scalable data analytic solutions, our goal is robust integration of machine learning with signal processing for measurement and control, in order to conduct research fundamental to large-scale, real-world questions in risk and liability management.
RiskEcon® Lab enables, facilitates and coordinates academic research focusing on these patterns and trends, through the development of commercially-viable, analytic applications employing computational statistical tools in conjunction with innovative and non-traditional data structures. In addition, the Lab’s activities involve the advancement of applied mathematical statistics and computational economics, through interdisciplinary post-doctoral, postgraduate, graduate research and education in data science and social computing in the public interest.