Dr. Martijn IJtsma

Dr. Martijn IJtsma

Assistant Professor at the Ohio State University

Current Projects

NSF CAREER Award: Making Robots More Cooperative Agents: Controlling Costs of Coordination Through Graph-Based Models of Joint Activity

The deployment of smart robots promises increased safety, productivity, and capability in domains such as disaster and emergency response, ground mobility, manufacturing, aviation, and space operations. Good human-robot collaboration is key to the realization of these promises. This project develops novel modeling techniques for analyzing and designing collaborative behavior in human-robot teams. Collaborative behavior requires adjusting and communicating with each other, thereby benefiting from participating in collaboration despite the cognitive and temporal costs of needing to coordinate with others. Such costs in human-robot collaboration can be high, as coordination with autonomous agents generally is more taxing and time-consuming than collaboration with other humans. The models developed as part of the award will help uncover the causes and effects of coordination costs in human-robot systems. Based on these models, the project develops techniques for managing coordination costs to avoid overloading human operators. Improved management of coordination costs will lead to more robust and resilient human-robot operations, broader adoption of smart robotic technologies, and realization of their promised benefits in a range of domains that are key to social welfare and national security. The project integrates education and outreach activities into the research for training the future workforce in systems thinking and interdisciplinary problem-solving skills. These skills will ready future engineers, researchers, and scientists to create integrated solutions in interdisciplinary environments to address complex engineering challenges that span technological, human, ecological, economic, and policy dimensions, among others.

FAA: Reliance on Automated or Complex Flight Deck Systems in Commercial Aircraft

This project’s objective is to develop methods for identifying and evaluating design characteristics of automated or complex system. Complex systems such as commercial aircraft, have many high vulnerabilities that can arise. These vulnerabilities may undermine flight crew performance when non-normal and abnormal events occur. The scope of this project is to provide the Federal Aviation Administration (FAA) with recommendations regarding the evaluation process of flight deck systems.

NASA: Contingency Planning for Advanced Air Mobility (AAM)

This project, led by Mosaic ATM, works on building a Contingency Planning Toolkit for Advanced Air Mobility. Our role is to develop and apply novel modeling and simulation techniques for evaluating candidate architectures and identify requirements for distributed AAM contingency planning. Through fast-time simulation, we can identify what is necessary to (1) create feasible concepts of operations that create the desired outcomes, (2) conduct comparisons between different system architectures and strategies for responding to contingencies, and (3) identify where further resources need to be invested to improve the robustness of envisioned concepts.

NASA: Cognitive Systems Engineering (CSE) Methods to Support Adaptive, Integrated Anomaly Response

The project, led by Applied Decision Sciences, aims to tailor cognitive systems engineering methods to support the design of Fault Detection, Isolation, and Recovery (FDIR) capabilities intended for remote smart habitats and NASA’s Moon to Mars initiative. We are adapting the Integrated Cognitive Analysis for Human-Machine Teaming (ICA-HMT) strategy, developed and exercised previously as part of the Army’s Future Vertical Lift program to understand the envisioned world for future rotorcraft that leverages advanced automation and operates in dynamic, time-critical and high-risk contexts. For this project, we are conducting an in-depth Cognitive Task Analysis (CTA) to understand and document the envisioned world of remote smart habitats and identify cognitive requirements associated with deep space operational challenges such as intermittently occupied smart habitats, limited communications bandwidth, and significant communication latencies. CTA findings are used to generate design recommendations for integrated human-autonomy system configurations that can then be evaluated using Work Models that Compute (WMC), a modeling and simulation framework. WMC is designed to model elements of collective work such as workload, interdependencies, and tradeoffs, incorporating macrocognitive aspects of the work that are not easily observed (e.g., sensemaking, decision making). We further design visualizations that enable developers to “what-if” and explore tradeoffs between different teaming configurations. Phase I objectives include identifying high-consequence FDIR use cases, extending WMC to include the cognitive aspects of the work that enable modeling of various integrated human-autonomy teaming

99P: Human-AI Teaming in Future Mobility

This collaboration with 99P Labs envisions new techniques for designing human-AI systems for future mobility. These techniques leverage ideas from functional modeling and graph theory to systematically reason about how different configurations of humans and AI capabilities create interdependence relationships that must be managed and supported through design. The project applies these techniques to the problem of replanning while en route, to envision how new machine capabilities can integrate with human cognitive processes around replanning to create a well-performing, integrated system.