Apr 2023
Speaker 1: Dr. Jun Chen
Title: Towards Heterogeneity in Multi-Target Tracking Using A Distributed Team of Mobile Robots
Abstract:
Multiple target tracking (MTT) is a fundamental problem in robotics,
wherein one or more robots simultaneously estimate the states of a
potentially large number of individual objects. These objects can be
either static or dynamic, and may leave or enter the task space over
time. The applications of MTT range from environmental monitoring to
surveillance to smart cities. In recent years, MTT problems have been
increasingly being solved by distributed mobile robotic networks due to
their capability to reactively adjust sensor coverage over the mission
space as new information is collected. Allowing a heterogeneous team of
robots to cooperate and heterogeneous targets to be effectively tracked
is an essential research topic to broaden the application of multi-robot
multi-target tracking (MR-MTT) in the real-world. This talk presents
our previous and ongoing work on the heterogeneity in MR-MTT in
threefold. Firstly, our work allows robots to simultaneously track
multiple classes of targets despite measurement uncertainty, including
false positive detections, false negative detections, measurement noise,
and target misclassification. Secondly, the heterogeneity of targets’
spatial distribution is considered, and algorithms are developed to
enhance the efficiency of target search and tracking. Thirdly, we
address the MR-MTT problems in complex scenarios where various types of
robots and sensors are needed to complete the task. Experimental and
simulated results are performed to demonstrate the promising efficacy
and application prospects of our proposed methods.
Bio:
Jun Chen received his M.S. degree in Electrical and Computer Engineering
from Stevens Institute of Technology in 2017 and his Ph.D. degree in
Mechanical Engineering from Temple University in 2021. He is currently a
postdoctoral fellow in the RISC Lab at KAUST. His research interests
include multi-robot system, information-based control, and robot
learning.
Speaker 2: Eesaa Harris
Title: Combining remote sensing and in situ data to understand mesoscale processes in the Red Sea
Abstract:
Ocean processes are complex and multidimensional and are thus
investigated in relative isolation for simplification. However, the
physical, chemical and biological components of the ocean are
intrinsically linked, and thus require novel approaches in order to gain
an understanding at the systems level. Furthermore, mesoscale processes
and features play an important role in regulating marine (and
atmospheric) systems, as their sum contribution have been shown to have
impacts at the global scale. The Red Sea provides a unique opportunity
to study these features, as it forms a “natural laboratory” relatively
isolated from the global oceans. Moreover, current climate trends
predict that various marine environments will end up in states similar
to that of the Red Sea. Hence, the objectives of the Integrated Ocean
Processes (IOP) lab at KAUST include understanding the overall
functioning of the Red Sea via integrated and contextual approaches that
investigate the linkages and cause-and-effects between different
components of the marine system. By combining both remote sensing and in
situ data, the relationships between physical components and their
biogeochemical responses in the Northern Red Sea (NRS) are investigated.
The area of sampling is characterized by dynamical structures at the
air-sea interface and within the water column, which was shown to
promote both the proliferation of phytoplankton as well as their
subsequent subduction into the ocean interior. This case study
highlights the limitation of temporal and spatial scales, and the need
to increase both in order to extrapolate to the entire Red Sea over
larger time scales.
Bio:
Eesaa is a first year PhD student in the Lab of Prof. Burton Jones. He
received his Master Degree in Oceanography at the University of Cape
Town, South Africa. His interests lay in biogeochemistry and the
relationships between physical and biogeochemical processes in tropical
oceans.