This seminar will feature Shiraz Khan,  Postdoctoral Researcher in the Robotics and Control Lab from the Department of Mechanical Engineering.

Title:

Error Identification and Correction in Sensor Networks

Abstract:

Groups of autonomous agents (e.g., UAVs and rovers) can be collectively deployed as a sensor network to achieve various objectives, such that the capabilities of the network far exceed those of the individual agent. A practical challenge which arises in the operation of sensor networks is the presence of faults, biases, or adversarial attacks, which can undermine the security and performance of the network. This talk will discuss the problem of error identification and correction (EIC) in sensor networks, wherein the objective is to identify the agents that have erroneous state estimates and reconstruct their errors. I will present a general framework for reconstructing errors by processing nonlinear measurements obtained across the sensor network, based on the assumptions that the errors are sparse (i.e., only a fraction of the agents have errors) and that the sensor network is sparsely connected. This two-fold sparsity of the EIC problem can be leveraged to yield a distributed EIC algorithm which may be interpreted as a multi-agent residual testing mechanism. The effectiveness of the sparsity-based EIC approach is conveyed using an example of sensor network localization in which the distances between connected agents are used to identify and correct the agents with erroneous estimates. The talk will briefly touch upon other estimation and control problems that arise in the multi-agent setting.

Bio:

Shiraz Khan earned his PhD in Aeronautics and Astronautics from Purdue University in 2023, following his Bachelor's degree in Aerospace Engineering from IIT Madras in 2018. He is presently a postdoctoral researcher in Mechanical Engineering at the University of Delaware. During his doctoral studies, he investigated the problems of robust and cyberattack-resilient state estimation in single and multi-agent systems. A recurring theme in his research is the investigation of the underlying structures (such as sparsity, topology, or symmetry) in various control theory and signal processing applications, and designing algorithms that can leverage these structures.