My research and project interests are guided by past graduate research and my teaching commitments.
Graduate Research Work
My graduate research was in the area of Automated Design, where I worked on Knowledge Representation (using graph grammars) of Planar Mechanisms and implemented an Optimization-based method to generate multiple linkages based on user requirements. This was primarily a position-based analysis and involved development of a Kinematic Analysis tool named PMKS (Planar Mechanism Kinematic Simulator). I was advised by Prof. Matthew I. Campbell (currently at Oregon State University) and Prof. Ashish Deshpande (at the University of Texas at Austin for 1 year)
Shown above is a snapshot of the overall implementation. A user sketches the profile that is required to be traced by a linkage. The optimization-based tool generated different linkage options in a reasonable amount of time. In addition to showing the potential of an automated design process, the goal was also to showcase the usefulness in generating multiple designs for the same problem. This would be ideal in a variety of different situations and unlike a manual process whereby generating such multiple solutions will take a long time. Shown below is a snapshot of the results generated by the tool for benchmark problems.
The PMKS tool’s UI development was spearheaded by Prof. Matthew I. Campbell using Microsoft Silverlight in Visual C#.
[For more info about this work, feel free to Contact me]
Insights from Teaching
At WPI, as I started my full-time teaching career and began teaching various courses, I realized the following:
- Availability of qualified teaching assistants is always a challenge. Teaching assistants (TAs) are always in demand. Besides, most graduate students are not necessarily familiar in all topics and find it difficult to learn and simultaneously provide quality feedback to students (there have been very good TAs/graders but hard to come by). Quality feedback to students is very important and given the increasing enrollment and workload, this takes a backseat and so does the student learning. Similar issue plagues online delivery of course content.
- Generic software tools (solid modeling and analysis) do not provide students with design alternatives and analysis insights, such as whether something is correct or not and what changes needed to be done for improvement. They also do not correspond to the analyses being taught in various courses because those software tools are generally a melange of various methods. Besides, students should be able to quantify performance characteristics through analytical means and make comparisons with software results.
- Manufacturing and assembly are aspects that are consistently overlooked at various universities due to the emergence of new techniques like 3D Printing. But there are a lot of machines that are still an amalgamation of multiple assembled parts. Students with no experience find it difficult to prototype quality systems. There are also no means to provide feedback on their designs.
- Commercial tools are expensive and are too specific for courses with low-enrollment. If some of the requirements listed earlier are important, then a lot of tools do not satisfy those requirements either.
- There are a lot of research developments happening in different labs across campus and the world. A lot of those developments do not necessarily percolate into courses at various levels. It is important for students to be exposed to those developments. For instance, the common refrain I hear from students enrolled in Mechanical Engineering is that they do not want any part of programming or controls systems but those are integral in design and manufacturing. But, it is important for all students to be exposed to different fields without overwhelming them.
My research themes are centered around addressing the challenges listed above and ensuring that students have an elevated learning experience. The projects are centered around the following themes:
- Automated Design and Manufacturing
- Entertainment and Medical Engineering
- Kinematics, Dynamics and Design Education
- Optimization, Machine Learning and Software Development
All the projects are currently handled through the undergraduate Major Qualifying Projects (MQPs), undergraduate and graduate Independent Study Projects (ISPs) and Master’s Thesis. MQPs are group activities while ISPs and Master’s Thesis activities are individual activities.
All the projects adhere to the following characteristics:
- Be generic so that the design can be easily customized to suit different requirements
- Adopt low cost and easily accessible techniques (for instance, use of 3D printing and open-source tools)
- Adhere to strict analysis methods using analytical and/or numerical techniques to validate design and performance
- Develop modules that can be easily adapted as course projects and modules to expose students to new developments in different fields
Automated Design and Manufacturing
In Automated Design and Manufacturing, my focus is to develop tools and techniques to automatically synthesize designs and generate manufacturing plans based on user specifications along varied scales (macro,micro,nano). Not only this, the plans are fed into different machines to automatically produce the parts and work with an integrated assembly system to assemble the devices. The ultimate aim is to develop demonstrable products that have real uses.
Within this broad area, I am also exploring the use of different techniques and the outcomes from various projects in other areas of interest to personalize and automate aspects of mechanical engineering education in the form of virtual labs.
Projects in the areas listed below are knowledge gathering exercises that will eventually aid projects in the area of Automated Design and Manufacturing. Not only that, this will be the umbrella theme that uses my graduate research work and addresses the challenges listed above.
Entertainment and Medical Engineering
My focus here is to develop devices by applying fundamental and innovative techniques in design and manufacturing. The projects currently in progress fall into two broad categories namely Bio-mimicking and Rehabilitation.
As the name suggests, Bio-mimicking involves developing customizable devices that are used to mimic various human and animal features. Such devices can be used to explain the working of various features and can be used in a variety of settings including medical education and rehabilitation studies. At the same time, the intricacies and the design and manufacturing challenges involved lead us to new research activities that are required for further developments in this area.
In terms of rehabilitation, the goal is to develop devices that can be used to rehabilitate humans and animals. Rehabilitation addresses different parts of the body and not just externally moving features.
Kinematics, Dynamics and Design Education
The design and manufacture of various machines rely on the solid fundamentals in Kinematics and Dynamics.
Optimization, Machine Learning and Software Development
Feel free to browse over different