Research Areas: Human-Centered Automation, Robotics, Decision Support Systems, Physics-Informed Artificial Intelligence, Computer-Aided Design, and Smart Manufacturing.
I am interested in developing human-centered automation solutions to increase human productivity, reduce human health risks, and enable innovation. Realizing human-centered automation requires robotics and automation technologies to reduce the physical human effort and enable humans to focus their effort on high-value tasks. It also requires augmenting human decision-making capabilities to reduce the probability of making mistakes, enabling creativity, and increasing decision-making speed.
I believe that robots can serve as smart assistants for humans. Many emerging robotics applications require multiple collaborating robots to operate under human supervision. To be useful in such applications, smart robotic assistants will need to: (1) program themselves, (2) efficiently learn from the observed performance, (3) safely operate in the presence of uncertainty, (4) appropriately call for help during the execution of challenging tasks, and (5) effectively communicate with humans.
We are making advances in physics-informed artificial intelligence to enable robots to exhibit smart behaviors. We are developing methods to automatically generate near-optimal trajectories in real-time to enable robots to program themselves from task descriptions. We are also developing self-supervised learning methods to equip robots with the ability to learn from observing the performance of previously executed tasks. We are developing methods for robots to operate safely in the presence of uncertainty by generating contingency-aware plans. We are developing computational methods for endowing robots with introspective capabilities so that they can seek help from humans on challenging tasks. If, during the execution, a contingency situation is detected, the system can issue appropriate alerts to humans and performs the necessary replanning. The system will assess its own confidence in executing a task, and if the risk appears to be high, it will seek human help. This approach enables the system to ensure human safety and enables it to quickly recover from errors. We are also exploring the use of augmented reality-based interfaces for enabling robots to elicit human guidance.
To augment human decision-making capabilities, I am interested in developing decision support systems that enable humans to make informed decisions in a timely manner for challenging applications. First, we are developing decision-making approaches that combine heuristic-aided discrete state-space search, non-linear optimization, and surrogate modeling to recommend informed decisions by the required decision-making deadlines. We are making advances in physics-informed artificial intelligence to enable the realization of decision support systems for applications that involve model uncertainty, complex physics, and fast decision-making speeds. Finally, we are developing a novel framework to combine model-based approaches and data-driven approaches in a consistent and unified manner. This enables us to exploit prior knowledge and augment missing components of models through safe and efficient learning.
Many applications will utilize multi-robot teams to perform environmental monitoring and information gathering tasks to provide situational awareness. Multi-robot systems can be particularly useful for dangerous missions in complex environments and reduce the risk to humans. Complex missions require the execution of spatially separated tasks by a team of robots. Planning strategies for such missions must consider the formation of effective coalitions among available assets and the assignment of tasks to robots with the goal of minimizing the expected mission completion time. The occurrence of unexpected situations that adversely interfere with the execution of the mission may require the execution of contingency tasks so that the originally planned tasks may proceed with minimal disruption. The planner must update the mission execution plan based on the probability of the potential contingency task to impact the mission tasks. It may either immediately incorporate the contingency task in the mission plan or defer its incorporation until the situation changes. We are developing a physics-informed artificial intelligence approach to proactively manage contingencies.
We anticipate that humans will be in supervisory roles to make decisions in critical situations and manage resources. At a high-level, humans will ensure that robots are working on tasks that align with the mission objectives as the mission progresses. In a mission with considerable uncertainty due to intermittent communications, degraded information flow, and failures, humans need to assess both the current and probable future states to make sound decisions. We are developing a simulation-based alert system that proactively notifies the human supervisor of possible undesirable events and serves as a decision support system for humans to make informed decisions.
Traditionally robots are used only on mass production applications. The manual programming of robots is economically not viable in high-mix applications. Therefore, many processing operations rely on manual labor in high-mix applications. The advent of human-safe robots is enabling robots to collaborate with humans on ergonomically challenging tasks and amplify human productivity. This enables robots to perform a large fraction of the task and only requires humans to perform the final touch-ups. In addition, the availability of 3D vision and force sensors enables robots to operate without custom fixtures and accommodate part and fixture variability. Smart robotic assistants powered by physics-informed artificial intelligence technology can program themselves from the high-level task descriptions and utilize sensor data to adapt their behaviors to deliver efficient and safe operational performance in high-mix applications. We are developing smart robotic assistants for a wide variety of manufacturing applications such as assembly, composite prepreg layup, kitting, finishing, inspection, and machine tending.
We are developing robotic cells to significantly expand additive manufacturing (AM) processes capabilities by enabling material deposition on non-planar geometries. Many composite parts have thin three-dimensional shell structures. Achieving the right fiber orientation is critical to the functioning of these parts. Printing them using conventional planar-layer AM processes leads to fibers being oriented in the plane of the layer. The capability to deposit the material using non-planar conformal layers can produce parts with improved material properties. Robots can be used to perform multi-resolution printing that finds the best trade-off between build speed and surface finish. Robots can also be used to realize supportless AM. AM is not expected to produce high-quality electronics in the near foreseeable future. Therefore, robots also enable the insertion of externally fabricated components such as sensors, actuators, and energy harvesting components during the AM process. Performing material deposition with robots requires solving many computational challenges. We are developing physics-informed artificial techniques needed for generating and executing robot trajectories to build high-quality parts using AM.