Autonomous Systems Initiative

Inaugural Workshop March 6th 2020

Donadeo Innovation Centre for Engineering, University of Alberta Edmonton, Alberta, Canada

Supported by

Inaugural Autonomous Systems Initiative (ASI) Workshop Friday March 6th 2020

8:30 am – 4:00 pm 8th Floor Donadeo Innovation Centre for Engineering

University of Alberta, Edmonton, Alberta

1

Agenda

Morning

Time Event Presenter/Participants

8:30 am Registration DICE 8-255ZZ All

9:00 am Opening Remarks DICE 8-292 Fraser Forbes

9:15 am Invited Talk: Scalable Predictive Knowledge Acquisition

Adam White

9:40 am Control and Monitoring Tools for Complex and Networked Autonomous Systems

Tongwen Chen

9:55 am Automated Transportation System with Enhanced Vehicle Intelligence and Infrastructure Digitization

Tony Qiu

10:10 am Autonomous Systems for Supporting Infrastructure Resilience and Response to Disasters

Amy Kim

10:25 am Autonomous Road Vehicles Systems for Improving the Emissions Footprint of Urban Passenger Transportation

Bob Koch

10:40 am Break All

10:55 am Autonomous and Semi-autonomous Systems for Healthcare Delivery

Mahdi Tavakoli

11:10 am Autonomous Systems for Industrial Communities

André McDonald

11:25 am Intelligent Technology Stack for Autonomous Systems

Henry Leung

11:40 am Closing Remarks Tony Qiu

11:50 pm Lunch / Student Poster Session DICE 8-255ZZ All

Inaugural Autonomous Systems Initiative (ASI) Workshop Friday March 6th 2020

8:30 am – 4:00 pm 8th Floor Donadeo Innovation Centre for Engineering

University of Alberta, Edmonton, Alberta

2

Agenda

Afternoon

(Researchers Only)

Time Event Presenter/Participants

1:15 pm Researcher Meeting DICE 7-395 Qiu/Project Researchers

1:45 pm

Project breakout sessions Project Researchers

Theme 1 - Methodologies & Tools for Autonomous Systems Project 1A: DICE 8-246

Chen, Marquez, Zhao

Theme 1 - Methodologies & Tools for Autonomous Systems Project1B: DICE 8-244

Leung, Reformat, Musilek

Theme 2 - Mobile Communities: DICE 8-222 Qiu, Zhang, Kattan, Buro, El-Basyouny, Kwon, Zaal, Far

Theme 3 - Sustainable Communities: Projects 3A/3B: DICE 8A-238

Kim, Koch, Lynch, Moussa, Whitehead, Qiu

Theme 4 – Healthy Communities DICE 8A-242 Tavakoli, Rouhani, Zheng, Lou, Tata

Theme 5 – Industrial Communities: DICE 8-238 McDonald, Cafferata, Ahmad, Hogan, Jägersand,

2:45 pm Break Project Researchers

3:00 pm Project breakout sessions continued Project Researchers

4:00 pm End of workshop

3

Oral Presentations 9:15 am – 11:50 pm Donadeo Innovation Centre for Engineering Scalable Predictive Knowledge Acquisition 9:15 am -9:40 am

Invited Talk: Adam White, Department of Computing Science, University of Alberta Adam White is an Assistant Professor at the University of Alberta and a Senior Research Scientist at DeepMind. He is a principal investigator of the Alberta Machine Intelligence Institute and the Reinforcement Learning and Artificial Intelligence group at the University of Alberta. Adam is a Canada CIFAR Chair in Artificial Intelligence. His work has been published in top conferences in ML and AI, including NeurIPS, ICLR, AAAI, IJCAI, and AAMAS. In particular, his work on off-policy learning and predictive knowledge representations is used by several major research groups as a foundation for their AI research.

Abstract: Intelligent systems like humans and animals know a lot about their world. People know that the sun will rise everyday, how to predict the trajectory of a ball in flight, and that their hand will burn if they touch a hot stove. These facts can all be phrased as predictive statements about our subjective future experience, conditioned on some way of behaving: look east to the horizon in the morning, visually tracking a moving object, and reaching out and touching the stove. The great strength of subjective prediction is that learning is embarrassing parallel: the system can scale the amount that it knows with increases in compute and data. The potential scalability of predictive learning has been known for years, however, prior demonstrations were limited to small numbers of predictions in small handcrafted toy problems. In this talk I will discuss a massively parallel prediction-learning system call Horde, based on ideas and algorithms from reinforcement learning. I will discuss the main strengths and limitations of Horde, and highlight how algorithms based on off-policy learning allow further parallel scaling in prediction learning. I will discuss how learning systems based-on Horde have achieved impressive results in visual navigation tasks and high degree-of-freedom robot-arm control. Combining Horde with large-scale compute and data provides one potential path towards AI systems that can know as much about the world as people and animals do.

4

Control and Monitoring Tools for Complex and Networked Autonomous Systems 9:40 am – 9:55 am

Project Lead: Tongwen Chen, Department of Electrical and Computer Engineering, University of Alberta Tongwen Chen is currently a Professor and Tier 1 Canada Research Chair in Intelligent Monitoring and Control at the University of Alberta, Canada. He received the BEng degree in Automation and Instrumentation from Tsinghua University (Beijing) in 1984, and the MASc and PhD degrees in Electrical Engineering from the University of Toronto in 1988 and 1991, respectively. His research interests include computer and network based control systems, event triggered control, process safety and alarm systems, and their applications to the process and power industries. He is a Fellow of IEEE, IFAC, as well as Canadian Academy of Engineering.

Project Team: Tongwen Chen, Horacio Marquez, and Qing Zhao (University of Alberta) Abstract: The motivation is two-fold: First, autonomous systems are often complex and interconnected by embedded computing devices and wireless networks, operating in a distributed, event driven fashion. Such systems call for new event driven control and monitoring methodologies. Second, most industrial facilities are operated by human operators through alarm monitoring systems. Could industrial plants be fully automated, without intervention of operators at all? Alarm systems are also event driven. This project focuses on three areas: Event driven control: How to design controllers and event triggering mechanisms for closed-loop stability and performance? Multi-agent systems: How to design event driven protocols which are implementable with distributed computing devices, achieving desirable global behavior? Industrial alarm monitoring: How to analyze and design practical alarm systems based on historical alarm and process data, developing advanced alarm analytics tools? How to automate industrial health monitoring, providing autonomous control and operational decisions? In this short presentation, the project motivation, scope of work, technical challenges, and impact of results will be reviewed, with some preliminary research findings shown at the poster session in the workshop.

5

Automated Transportation System with Enhanced Vehicle Intelligence and Infrastructure Digitization 9:55 am – 10:10 am

Project Lead: Tony Qiu, Department of Civil and Environmental

Engineering, University of Alberta

Tony Qiu is an Associate Professor in the Department of Civil and Environmental Engineering at the University of Alberta (U of A) and holds both the Canada Research Chair in Cooperative Transportation Systems and the NSERC Industrial Research Chair in Intelligent Transportation Systems. Since joining the U of A in 2009, Dr. Qiu founded the Intelligent Transportation Systems (ITS) research lab, now called the Centre for Smart Transportation (CST). He is also the Scientific Director for the Autonomous Systems Initiative (ASI), a multi-million dollar, Campus Alberta, research program focused on developing Artificial Intelligence and automated

systems. Dr. Qiu received his Ph.D. from the University of Wisconsin-Madison in 2007 and his BSc. and MSc. from Tsinghua University of China in 2001 and 2003, respectively. From 2008-2009, he worked as a postdoctoral researcher in the California PATH Program at the University of California, Berkeley. Project Team: Tony Qiu, Hong Zhang, Randy Goebel, Michael Buro, Karim El-Basyouny, Tae Kown (University of Alberta), Behrouz Far, Lina Kattan (University of Calgary), Chris Zaal (SAIT) Abstract: Theme 2's research focuses on the future digitization of transportation infrastructure to better support the movement of people and goods. As information and communication technology rapidly advances, there is no better time to re-think our fundamental knowledge about roads and vehicles. Our research team aims to develop an autonomous transportation system with enhanced intelligence of both vehicles and the infrastructure in which they run, to assist in the progress of mobility and goods movement and to address the growing transportation challenges within our communities.

6

Autonomous Systems for Supporting Infrastructure Resilience and Response to Disasters 10:10 am – 10:25 am

Project Lead: Amy Kim, Department of Civil and Environmental Engineering, University of Alberta Amy Kim is an Associate Professor of Transportation Engineering, in the Department of Civil & Environmental Engineering at the University of Alberta. Her research program in transportation systems analysis focuses on understanding multimodal networks and applications for resource allocation across long-distance, large-scale systems. She is interested in wildfire emergency management, northern transportation infrastructure planning, and integrated analysis of air-ground transportation. She joined the U of A faculty in 2011 after completing her PhD at the University of California, Berkeley. In the four years between

her MSc (2002, Berkeley) and PhD studies, she worked in transportation engineering consulting in both Oakland, CA and Vancouver, BC. Project Team: Amy Kim, Bob Koch, Alan Lynch, Walied Moussa (University of Alberta),

Ken Whitehead (SAIT), Chris Hugenholtz (University of Calgary)

Abstract: This theme focuses on the development of a knowledge platform for Autonomous Systems in developing capacities to reduce the environmental impact and consequences of disaster events. The vision is towards regular, safe deployment within ecological, industrial, and urban settings, with a focus on environmental and social benefits. Examples include coordinated aerial systems for inspection of ageing infrastructure and capabilities for autonomous disaster response and recovery, including technology integration into emergency response management.

7

Autonomous Road Vehicles Systems for Improving the Emissions Footprint of Urban Passenger Transportation 10:25 am – 10:40 am

Project Lead: Bob Koch, Department of Mechanical Engineering, University of Alberta C.R. Koch received his B.S. degree in mechanical engineering from the University of Alberta, Edmonton, Canada in 1985, and his M.S. and Ph.D. degrees from Stanford University, Palo Alto, CA, in 1986 and 1991, respectively. From 1991 to 1992 and from 1994 to 2001 he worked at Daimler-Benz DaimlerChrysler in Stuttgart Germany in advanced internal combustion engines. During 1992 to 1994 he worked for General Motors. In 2001, he joined the Mechanical Engineering Department of the University of Alberta, Edmonton Canada, where he is a Professor. His research interests include combustion engines,

sensors for emission measurement during driving, advanced powertrains and control of fluid systems. He is currently an Associate Editor for the Mechatronics Journal (Elsevier). Project Team: Bob Koch, Amy Kim, Tony Qiu (University of Alberta) Abstract: This theme focuses on the development of a knowledge platform for Autonomous Systems in developing capacities to reduce the environmental impact and consequences of disaster events. The vision is towards regular, safe deployment within ecological, industrial, and urban settings, with a focus on environmental and social benefits. Examples include autonomous road vehicles systems for improving the environmental footprint of passenger transportation.

8

Autonomous and Semi-autonomous Systems for Healthcare Delivery 10:55 am- 11:10 am

Project Lead: Mahdi Tavakoli, Department of Electrical and Computer Engineering, University of Alberta Mahdi Tavakoli is a Professor in the Department of Electrical and Computer Engineering, University of Alberta, Canada. He received his BSc and MSc degrees in Electrical Engineering from Ferdowsi University and K.N. Toosi University, Iran, in 1996 and 1999, respectively. He received his PhD degree in Electrical and Computer Engineering from the University of Western Ontario, Canada, in 2005. In 2006, he was a post-doctoral researcher at Canadian Surgical Technologies and Advanced Robotics (CSTAR), Canada. In 2007-2008, he was an NSERC Post-Doctoral Fellow at Harvard University, USA. Dr. Tavakoli’s research

interests broadly involve the areas of robotics and systems control. Specifically, his research focuses on haptics and teleoperation control, medical robotics, and image-guided surgery. Dr. Tavakoli is the lead author of Haptics for Teleoperated Surgical Robotic Systems (World Scientific, 2008). He is a Senior Member of IEEE and an Associate Editor for IEEE/ASME Transactions on Mechatronics, Journal of Medical Robotics Research, Control Engineering Practice, and Mechatronics. Project Team: Mahdi Tavakoli, Hossein Rouhani, Bin Zheng, Edmond Lu (University of Alberta), Garnette Sutherland (University of Calgary), Matthew Tata (University of Lethbridge) Abstract: The objective of the research project of the Healthy Communities Theme is to develop new intelligent and autonomous system technologies for robot-assisted surgery, therapy and rehabilitation that will reduce the burden on Canada’s and Alberta's health care system by making medical interventions more efficient, accurate, accessible, and reliable. Intelligence and autonomy in healthcare settings require utilizing the advantages offered by robots and the real-time decision-making capabilities of machines. To this end, this project builds on and brings together Alberta’s strengths in intelligent, wearable and robotic systems for medical and biomedical applications.

9

Autonomous Systems for Industrial Communities 11:10 am – 11:25 am

Project Lead: André McDonald, Department of Mechanical Engineering, University of Alberta André McDonald is a Professor and the Associate Chair (Research) in the Department of Mechanical Engineering at the University of Alberta (UofA). He has more than fifteen years of experience in the fabrication, development, and performance assessment of thermal- and cold-sprayed coatings fabricated with fixed robots. His current research involves the development of flame-sprayed multi-functional coatings that provide wear and erosion resistance and heating and structural health monitoring to a variety of industry structures. He has received several awards including Fellow of The Institute of Materials, Minerals and Mining, Fellow of ASM,

the President’s International Fellowship Initiative Award from the Chinese Academy of Sciences, the Mentorship Award from the Faculty of Engineering (University of Alberta), the Composites Conference Best in Track Technical Paper Award for Manufacturing, and the Association of Professional Engineers and Geoscientists of Alberta’s Early Accomplishment Award. He is currently the Lead Editor of the Journal of Thermal Spray Technology, Chairman of the Canadian Cold Spray Alliance, and President of the ASM Thermal Spray Society Board. Project Team: André McDonald, Rafiq Ahmad, James Hogan, Martin Jägersand

(University of Alberta), Alicia Cafferata-Arnett (RDC)

Abstract: This theme in the Autonomous Systems Initiative will develop Autonomous Systems for advanced manufacturing and repair. It deals with safely protecting our environment, responsibly developing our natural resources, and efficiently manufacturing the products that we need to support and improve our quality of life. The theme will support industrial efforts to reduce environmental impact of disaster events. It will also utilize event-triggered control tools for Autonomous Systems in an on-site industry setting. In order to accomplish these goals, we will develop hybrid intelligent additive manufacturing enabled by semi-autonomous vision-guided robot navigation and manipulation. These solutions will be used for in situ in-field assessment and repair without on-site human intervention. By targeting extreme environments that are in corrosive, low and high temperatures, underground, mining, or remote locations, we will enable immediate robotic response and leaks/disaster mitigation/avoidance at remote unmanned infrastructures. These infrastructures may include well-heads, pumping stations, pipelines, electric sub-stations, and lines.

10

Intelligent Technology Stack for Autonomous Systems 11:25 am – 11:40

Project Lead: Henry Leung, Department of Electrical and Computer Engineering, University of Calgary Henry Leung is a professor of the Department of Electrical and Computer Engineering of the University of Calgary. Before joining U of C, he was with the Department of National Defence (DND) of Canada as a defence scientist. His main duty there was to conduct research and development of automated surveillance systems, which can perform detection, tracking, identification and data fusion automatically as a decision aid for military operators. His current research interests include big data analytic, chaos and nonlinear dynamics, information fusion, machine learning, signal and image processing, robotics and internet of things. He has published

extensively in the open literature on these topics. He has over 270 journal papers and 250 refereed conference papers. Dr. Leung has been the associate editor of various journals such as the IEEE Circuits and Systems Magazine, International Journal on Information Fusion, IEEE Trans. Aerospace and Electronic Systems, IEEE Signal Processing Letters, IEEE Trans. Circuits and Systems. He has also served as guest editors for the special issue “Intelligent Transportation Systems” for the International Journal on Information Fusion and “Cognitive Sensor Networks” for the IEEE Sensor Journal. He is the topic editor on “Robotic Sensors” of the International Journal of Advanced Robotic Systems. He is the editor of the Springer book series on “Information Fusion and Data Science”. He is a Fellow of IEEE and SPIE. Project Team: Henry Leung (University of Calgary), Marek Reformat, Petr Musilek (University of Alberta) Abstract: Our project develops tools and methods to support necessary capabilities for autonomous systems. I particular, we develop artificial intelligence (AI) methodologies for monitoring and control of modern systems in various communities such as 1) industrial communities: safer cities and infrastructure with IoT, 2) mobile communities: aerospace and intelligent transportation systems, 3) sustainable communities like pipeline in Alberta, 4) Healthy communities: wearable and health sensor monitoring. To address the needs in these communities, we focus on developing techniques on anomaly detection for multiple sensor systems, machine learning for edge and core computing, distributed control and sensing, and autonomous decision making. In this talk we present our research progresses on these methodologies and their applications to the above different communities.

11

Poster Presentations 12:00 pm – 1:15 pm Donadeo Innovation Centre for Engineering

Poster #

Presenter Affiliation Poster Title Theme

1 Hao Yu University of Alberta Event-Triggered Control of Networked Autonomous Systems

1

2 Bahador Rashidi University of Alberta Autonomous Root-Cause Fault Diagnosis

1

3 Shimin Wang University of Alberta Event-Triggered Synchronization of Autonomous Multi-Agent Systems

1

4 Mohsen Ghodrat University of Alberta Separation Principle for Event-Triggered Control Systems

1

5 Zheng Yu University of Alberta Fragmentation Coagulation Based Mixed Membership Stochastic Blockmodel

1

6 Marcin Pietrasik University of Alberta A Simple Method for Inducing Class Taxonomies in Knowledge Graphs

1

7 Elizaveta Kharlova University of Alberta Deep Learning Sequence to Sequence Model with Attention

1

8 Shuoyan Xu University of Alberta Connected Vehicle-Enabled Weather Responsive Snow and Ice Control Strategies

2

9 Yuwei Bie University of Alberta A CV-CACC Algorithm to Divide and Reform Connected Vehicle Platoons at Signalized Intersections to Improve Traffic Throughput and Safety

2

10 Xiao-Long Wang University of Alberta Baseline Experiments for Vision-based Collective Perception

2

11 Ehsan Ahmadi University of Alberta Collective Driving: A Distributed Pose-Graph Optimization Approach

2

12 Huiyu Chen University of Alberta System Optimal-based Joint Optimization of Traffic Flow and Signal Control in the Connected Vehicle Environment: A Model Predictive Control Method

2

13 Kaizhe Hou University of Alberta An RSE-Assisted Connection Restoration Method for Urban Arterial Roads in A Connected Vehicle Environment

2

12

Poster #

Presenter Affiliation Poster Title Theme

14 Andres Rosales University of Alberta The Implications of Light, Weather and Reflectivity Variations on Automatic Traffic Sign Recognition Performance

2

15 Fan Wu University of Alberta Mobility Pilot based Automated Intersection Control

2

16 Jiangchen Li University of Alberta Fast Adaptive Signal Control in the Connected Vehicle Environment

2

17 Can Zhang University of Alberta Decentralized Urban Traffic Signal Control Under Different Network Decompositions

3

18 Maja Kucharchyk University of Calgary Pre-disaster Mapping with Drones: An Urban Case Study in Victoria, BC, Canada

3

19 Mohammed Kayed

University of Alberta Critical Infrastructures Monitoring using Smart Autonomous Distributed Intelligent Micro Sensory

3

20 Amr Balbola University of Alberta Detection of landslides and avalanches using Autonomous Distributed Intelligent Micro Sensory

3

21 Sabrena Ohi University of Alberta Locating Emergency Shelter Infrastructures over the Alberta Highway Network

3

22 Amir Moeini University of Alberta Robust Motion Control of Quadrotor UAVs

3

23 Awais Rafique University of Alberta Vision-based Control of Unmanned Aerial Vehicles for Inspection and Maintenance Applications

3

24 Ali Lotfi University of Alberta Measurement of Cold Climate NOx Emissions of a Light Duty GDI Engine Vehicle using PEMS* in Real-world Driving Conditions

3

25 Vahid Hosseini University of Alberta Air Quality Research, An Integrated Approach from Emission to Health Impact

3

26 Lingbo Cheng University of Alberta Neural-Network-Based Heart Motion Prediction for Ultrasound-Guided Beating-Heart Surgery

4

13

Poster #

Presenter Affiliation Poster Title Theme

27 Ali Torabi University of Alberta Application of a Redundant Haptic Interface in Enhancing Soft-Tissue Stiffness Discrimination

4

28 Andi Liu University of Alberta A Smart IoT System to Fully Automatic Control Brace Loading for Scoliosis Treatment

4

29 Ahmed Badr University of Alberta A High Efficiency Energy Harvester Device to Power IoT Systems

4

30 Lukas Grasse University of Lethbridge

Auditory Awareness and Speech Control for Assistive Robotics

4

31 Alireza Noamani University of Alberta An Autonomous System for Postural Control Characterization after Incomplete Spinal Cord Injury

4

32 Milad Nazarahari University of Alberta Development of an Autonomous System for Objective Outcome Evaluation of Patients with a Shoulder Injury

4

33 Milad Nazarahari University of Alberta Application of Wearable Sensors in Remote Daily Activity Monitoring

4

34 Amirhossein Mahdavi

University of Alberta Numerical Study of Tensile Behaviour of Cold-Sprayed MMC Coatings

5

35 Milad Rezvani-Rad University of Alberta Microstructural and Performance Analyses of Thermally Sprayed Electric Resistance Heating Systems Used for Protection of Steel Pipes Against Bursting

5

36 Masood Dehghan University of Alberta Semi-autonomy in Tele-operation - Path Specification and Adaptation

5

37 Mohammad Parsazadeh

University of Alberta Development and Performance Characterization of Advanced Coatings

5

38 Mana Azizi University of Alberta An Innovative Approach Towards a Novel Autonomous Repair System for Damaged Components

5