The Workshop on
Data-Driven and Intelligent Cyber-Physical Systems (DI-CPS)

In conjunction with the
2021 CPS-IOT Week

Background

Smart Cities & Transportation CPS has evolved in recent years to produce a deluge of data.

These data comprise of safety connectivity, health services, transportation, and mobile nodes such as vehicles themselves as well as infrastructure like road-side units, DSRC units, and pole-mounted cameras, among others. Those data are extremely valuable in determining human behavior both at the scale of the entire population as well as at the level of individual people navigating the infrastructure landscape.

With the advent of IoT, sensor data is being generated at a pace and volume that is difficult to process and make inferences from. With several data modalities in the picture, new opportunities arise in terms of data collection, validation, analysis, inference making. These additional data allow for new efforts to develop novel models for traffic behavior at different geographical locations and time points, as well as what factors are consequential human driving behavior. At the same time, there is a growing need for automated applications to be fair, secure, and resilient. Participants in the workshop will exchange ideas on these and allied topics, including data science and open-source data sets for smart cities, decision making for smart cities, design of intelligent systems in smart cities, and challenges in deployment, equity, and fairness in smart cities, and security and privacy in AI for cities.

Scope

The Workshop on Data-Driven and Intelligent Cyber-Physical Systems (DI-CPS) invites academia, industry, and governmental entities to submit:

• Original research papers on the Transportation CPS concerning data with a human in the loop
• Software tools for integrative analysis of transportation gather from multiple modalities
• Demos (hands-on or videos) of experiences/testbeds/software/novel architecture of CPS and IoT transportation data research

We seek submissions from researchers across the domain of transportation, connected and autonomous vehicles, and smart cities tackling the issue of data-driven modeling to study human traffic behavior, the influence of autonomous vehicles on human traffic and vice-versa, and multi-modal traffic data analysis, including but not limited to:

• Novel and low-cost hardware and software infrastructure for seamless data gathering from onboard sensors
• CPS transportation data analysis and modeling tools
• Machine learning applications for integrative analysis of data from multiple modalities
• Tools for CPS Big data for Transportation
• Novel applications to human-in-the-loop transportation CPS

Further, of interest will be papers that can direct the research community to new research paradigms, and those that can set research agendas and priorities in transportation CPS data with a human in the loop. The deadline for paper submission is 14th March 2021.

Submission Guidelines

Submissions include long papers (8-12 pages) , short papers (4-6 pages) and concept papers (2 pages)

Long papers include:

1. Original research on transportation CPS data with a human in the loop
2. Integrative and multi-modal data analysis resulting in novel models for human driving behavior, and vehicular control in mixed autonomy.
3. Approaches to model complex decision-making tasks in smart cities and how they tackle uncertainty.
4. Challenges faced and lessons learned in deploying intelligent systems in smart cities.
5. Principled heuristics to design scalable decision-making in city-scale CPS.
6. Anomaly detection in smart and connected communities.

Short papers include:

1. Original preliminary research on transportation CPS data with a human in the loop
2. Demos, experience, tutorials, and presentation on software tools, simulations, and experimental results concerning CPS transportation with a human in the loop.
3. Frameworks and machine learning models for solving the challenge of integrating heterogeneous and cross-domain data in smart cities.

Submitted papers can be up to 12 or 6 pages including appendices and references. Submissions must use the format defined by ICCPS (http://iccps.acm.org/2021/call-for-paper/). Only PDF or latex-zipped files will be accepted. Alongside the camera-ready PDF’s, you will be required to collect and submit the source files for each paper – all files which were used to create the final output (PDF), be they Word, LaTeX, image files, etc. There is no requirement to anonymize the submissions; Authors may choose to submit anonymously or not. Accepted papers will be published by the ACM Press and/or the ACM Digital Library; however, authors can choose to opt out of formal proceedings. We welcome prior work published in conferences or journals. Each accepted paper must be presented by a registered author. Submissions not meeting these guidelines risk immediate rejection. For questions about these policies, please contact the chairs.

Please submit your work at https://easychair.org/conferences/?conf=cpsiotdata2020

Important Dates

• Title/Abstract Submission Deadline: February 10, 2021 Abstract will be accepted any time before the PDF update (23:59 Anywhere on Earth time)
• PDF Update Deadline: March 14, 2021 (23:59 Anywhere on Earth time)
• Notification of Acceptance/Rejection: April 1, 2021
• Camera-Ready Papers Due: (hard deadline): April 20, 2021

Program Chairs

• Katie Driggs-Campbell (University of Illinois, Urbana Champaign)
• Ayan Mukhopadhyay, (Vanderbilt University)
• Raphael Stern (University of Minnesota)
• Aron Laszka, (University of Houston)

Technical Program Committee

• Rahul Bhadani (The University of Arizona)
• Matthew Bunting (The University of Arizona)
• Will Barbour (Vanderbilt University)
• Xuan Sharon Di (Columbia University)
• Jnaneshwar Das (Arizona State University)
• Benjamin Seibold (Temple University)
• Daniel Work (Vanderbilt University)
• Abhishek Dubey (Vanderbilt University)
• Ram Rajagopal (Stanford University)
• Danny Huang (New York University)

Steering Committee

• Rahul Bhadani (The University of Arizona)
• Raphael Stern (University of Minnesota)
• Daniel Work (Vanderbilt University)

Web Chair

• Rahul Bhadani (The University of Arizona)