Joint Workshop on Interactive Prediction and Decision-Making on IV2020

Joint Workshops on IV2020:

From Benchmarking Behavior Prediction to Socially Compatible Behavior Generation in Autonomous Driving

Zoom link to the workshop (Nov. 6, 9am-5pm PST): https://zoom.us/j/95266012447?pwd=bCtRRmdMcG5NWk5NN0xqSDdWdFRmZz09

Workshop Description

In the past decade, extensive progress has been made in autonomous driving in both academia and industries, and more and more autonomous vehicles are deployed to drive on public roads. To negotiate its position in a space that is occupied by humans (e.g., human drivers, pedestrians), an autonomous vehicle need to effectively predict the human behaviors and generate socially compliant actions to enable efficient and safe interaction with humans. The performances of behavior perdiction and behavior generation are mutually influenced in a sense that an accurate prediction will help to generate better decisions and planning strategies, and an robust behavior generation strategy can compensate for uncertainties in the prediction results. This joint workshop aims to address fundamental problems in these two topics, and discuss in particular the connection/mutual influence between them.

The joint workshop has two themes:

1) Performance evaluation of prediction algorithms, particularly in the presence of specific planning algorithms;

2) Behavior generation algorithms under uncertainties from prediction.

For more detailed scopes and call for paper of each workshop, please check out the corresponding websites below.

Benchmarking Probabilistic Behavior Prediction in Highly Interactive Driving Scenarios

(Code: 2upm6)

deadline: March 14th, 2020

Behavior Generation and Decision Making for Socially Compatible Autonomous Vehicles

(code: 37tdg)

deadline: March 14th, 2020

Invited Speakers:

Christoph Stiller

(Professor at KIT)

Christoph Stiller is Full Professor and the Director of the Institute of Measurement and Control at KIT, Germany. His research interests include the full stack of automated driving, starting at multimodal sensor calibration, via localization and mapping, environment perception and scene understanding, to decision making and motion planning. He served as a President (2012–2013) and several other positions for the IEEE Intelligent Transportation Systems (ITS) Society. His Autonomous Vehicle AnnieWAY was a finalist in the Urban Challenge 2007 and the winner and second winner of the Grand Cooperative Driving Challenge 2011 and 2016, respectively. In 2013, he collaborated with Daimler on the automated Bertha Benz Memorial Tour.

Changliu Liu

(Professor at RI at CMU)

Dr. Changliu Liu is an assistant professor in the Robotics Institute at Carnegie Mellon University (CMU). Her research interests lie in the design and verification of intelligent systems with applications to transportation and manufacturing. She published the book “Designing robot behavior in human-robot interactions” with CRC Press in 2019. She is the recipient of the 2019 Amazon Research Award.

Katherine Driggs-Campbell

(Professor at UIUC)

Katie Driggs-Campbell is currently an assistant professor at the University of Illinois at Urbana-Champaign in the Department of Electrical and Computer Engineering. Prior to that, she was a Postdoctoral Research Scholar at the Stanford Intelligent Systems Laboratory in the Aeronautics and Astronautics Department. She received a B.S.E. with honors from Arizona State University in 2012 and an M.S. from UC Berkeley in 2015. She earned her PhD in 2017 in Electrical Engineering and Computer Sciences from the University of California, Berkeley. Her lab works on human-centered autonomy, focusing on the integration of autonomy into human dominated fields, merging ideas robotics, learning, transportation, and control.

Qi Luo

(Staff Software Architect at Baidu USA)

Qi Luo is Software Architect and Tech Lead manager for Baidu Apollo Platform, his team is responsible for the prediction, planning, control and vehicle interface models in Apollo Platform. Qi earned a Bachelor degree in Shanghai Jiaotong University in China and a PhD in Purdue University in Robotics and Automation, before join in Baidu, he worked in Rockwell Automation for Industrial Robotics Trajectory Planning and Control.

Nan Li

(Ph.D. at University of Michigan)

Nan Li is a final-year PhD student at the University of Michigan, Ann Arbor, working with Prof. Ilya Kolmanovsky and Prof. Anouck Girard. His research interests include control of constrained and stochastic systems, and modeling and control of multi-agent, human-machine systems.

Tentative Schedules (Nov. 6, Pacific Time):

9:00 am - 9:05 am: Welcome and Overview
9:05 am - 9:50 am: Professor Christoph Stiller (Dr. Maximilian Naumann on behalf), Safe and Convenient Behavior for Automated Vehicles
9:50 am - 10:20 am: Dr. Wei Zhan, Introduction to the INTERACTION dataset and prediction challenge
10:20 am - 10:30 am: break
10:30 am - 11:15 am: Professor Katherine Driggs-Campbell, Prediction Methods for Safe Vehicle-Pedestrian Interaction
11:15 am - 12:00 pm: Nan Li, Game-Theoretic Modeling of Traffic Scenarios Involving Multi-Vehicle Interactions and Applications
12:00 - 1:30pm: lunch break
1:30pm - 2:15 pm: Professor Changliu Liu, Prediction and Planning for safe and interactive autonomous driving
2:15 pm - 3:00 pm: Dr. Qi Luo, Apollo 6.0: What is under the hood?
3:00 pm - 3:30 pm: panel
3:30 pm - 4:00 pm: break
4:00 pm - 5:00pm: workshop papers