6 Dec
2023
6 Dec
'23
6:45 a.m.
*For all details see attachments, *
*The application will be open subject to the availability of
the positions *
*Position 1: *under Asian Smart Cities Research and Innovation Network
(ASCRIN) in collaboration with BITS Pilani and La Trobe University,
Melbourne Australia,
*Title of the project:* “Low-cost empathetic robotic solution for
caregiving (LERSCe)”.
*Abstract:* Australia and India have an elderly population which is rapidly
growing. Studies indicate that almost half of the current population of
over 75’s suffer from physical and/or mental impairments and as a result
are in need of a high level of care. With an increasing elderly population
and limited healthcare professionals, automated AI-aided health caregiving
services offer a scalable solution to cope. Moreover, these solutions can
be made efficient and cost-effective. One key service area where need
outstrips human resources is in the delivery of remote physiotherapy. This
project proposes to develop a novel soft-robotic actuator for the purpose
of delivering remote physiotherapy in the home. A closed loop edge AI
framework will drive the actuators to optimize service delivery. The edge
framework will be self-contained to address privacy and security concerns.
Control models will be trained offline and run locally on edge-capable
devices. Machine learning methods, such as deep learning, will be explored
to optimize actuator operations for maximum comfort and service efficacy.
The feedback loop may include facial and audio expressions and information
from other sensors such as skin resistance and
pressure to measure performance. Hence, there are two distinct aspects of
the project: 1) the development of a lightweight, low-cost soft-robotic
actuator 2) the development of an actuator control model which factors user
behaviour and emotional feedback. Feedback may be derived from a suite of
multimodal sensors including, but not limited to vision, tactile, and skin
resistance.
*Position 2: *funded by BITS Cross-disciplinary Research Funding (CDRF)
*Title: *Design, Development and Control of a low-cost AI-enabled 3-DOF
Compliant Manipulator
Abstract: The aim of this project is to develop a robotic manipulator
system that can be safely employed with human workers and which can
effectively coordinate with them. This would require an interdisciplinary
approach towards System Design, Control Law Design and Brain Computer
Interface (BCI). A 3-DoF robotic manipulator model with flexible joints and
torque controlled actuators will be designed. Then incorporate improved
user-machine adaptation algorithms in a BCI.
*Position 3 and 4: *funded by TiH-ISI Kolkata
*Title:* Generalized Tampering Detection in Media (GTDM)
Abstract: We plan to robustify and generalize the manipulation attack
detection by exploring monocular-based depth to better discriminate the
spatial and temporal synergy in the media. Along with this, we will explore
the power of SSL and multi-task learning (MTL) while learning the 3D
reconstruction.