Agenda for 3rd VeHICaL Project Annual Meeting

October 3, 2019 at UC Berkeley, HMMB (Hearst Mining) 290

October 4, 2019 at UC Berkeley, 540 Cory Hall

Day 1 (in 290 Hearst Mining)

8:00-8:30: Breakfast
8:30-8:40: Welcome & Introductions
8:40-9:00: Project Overview [Sanjit Seshia]

Session 1: Human Modeling and Interaction

9:00-9:20: Human-machine collaboration and information processing limitations [Mark Ho]
9:20-9:40: Modeling human attention as optimal sequential sampling [Fred Callaway]
9:40-10:00: Learning Task Specifications from Demonstrations [Marcell Vazquez-Chanlatte]
10:00-10:20: Maximum Likelihood Constraint Inference from Demonstrations [Dexter Scobee]
10:20-10:30: Nonverbal Robot Feedback for Human Teachers [Isabella Huang]

10:30-11:00: Break 

Session 2: Control and Verification

11:00-11:20: Safe Learning-Based Control [Claire Tomlin]
11:20-11:40: A Hamilton-Jacobi Reachability-Based Framework for Predicting and Analyzing Human Motion for Safe Planning [Andrea Bajcsy and Somil Bansal]
11:40-12:00: Counterexample-Guided Synthesis of Perception Models and Control [Hadi Ravanbakhsh]
12:00-12:20: Towards Assume-Guarantee Profiles for Autonomous Vehicles [Tung Phan and Karena Cai]
12:20-12:40: VerifAI: A Toolkit for the Formal Design and Analysis of AI-Based Systems [Daniel Fremont]

12:40-2:00: Lunch & poster session

Session 3: Human Interfaces, Education and Outreach

2:00-2:20: Education Efforts in AR/VR and Mobility [Bjoern Hartmann]
2:20-2:30: Girls in Engineering Summer Program  [Lizzie Hager-Barnard]
2:30-2:50: High Resolution Soft Tactile Sensor for Physical Human-Robot Interaction [Isabella Huang]
2:50-3:00: Optimal Design of Human-Machine Interfaces [Mark Ho]

Session 4: Visitor/Alumni talks

3:00-3:20: RuleBooks for Autonomous Vehicles [Nok Wongpiromsarn (nuTonomy)]
3:20-3:40: Fly-by-Logic: Trajectory Generation for Drone Fleets with Temporal Logic Objectives [Yash Pant (UPenn, new VeHICaL postdoc)]
3:40-4:00: Analyzing the Boeing 737 Max Incidents [Jaime Fernandez Fisac, alumnus]

4:00-4:20: Break

Final Session

4:20-5:00: Discussion & Feedback, plans for Day 2

Day 2 (in 540 Cory Hall)

9:00-10:30: Breakout sessions on the following topics:

  1. Human Modeling and Human-Machine Interaction
  2. Requirements and Causality
  3. Verified Machine Learning for Perception and Autonomy

10:30-11: Break

11:00-12:00: Report back and discussion

12:00-1:00: Lunch