- Speaker: Dr. Jianxiong Xiao, AutoX
Abstract: Today, there are two major paradigms for vision-based autonomous driving systems: mediated perception approaches that parse an entire scene to make a driving decision, and behavior reflex approaches that directly map an input image to a driving action by a regressor. In this paper, we propose a third paradigm: a direct perception based approach to estimate the affordance for driving. We propose to map an input image to a small number of key perception indicators that directly relate to the affordance of a road/traffic state for driving. Our representation provides a set of compact yet complete descriptions of the scene to enable a simple controller to drive autonomously. Falling in between the two extremes of mediated perception and behavior reflex, we argue that our direct perception representation provides the right level of abstraction. We evaluate our approach in a virtual racing game as well as real world driving and show that our model can work well to drive a car in a very diverse set of virtual and realistic environments.
- Panelists:
- Prof. Kang Shin (University of Michigan)
- Dr. Jianxiong Xiao (AutoX)
- Prof. Ashwin Ashok (Georgia State University)
- Prof. Shubham Jain (Old Dominion University)
- Moderator: Prof. Michael Tsai (National Taiwan University)
- Privacy Risks in Vehicle Grids and Autonomous Cars (invited paper, 20 min)
- Detecting False Position Attack in Vehicular Communications Using Angular Check (15 min)
Josh Joy (UCLA); Mario Gerla (UCLA);
Seungho Kuk (Korea University); Hyogon Kim (Korea University); Yongtae Park (Korea University)
- Vehicular Micro Clouds as Virtual Edge Servers for Efficient Data Collection (invited paper, 20 min)
- Characterizing the “Driver DNA” Through CAN Bus Data Analysis (invited paper, 20 min)
- CarLab: Framework for Vehicular Data Collection and Processing (invited paper, 20 min)
Florian Hagenauer, Christoph Sommer (Heinz Nixdorf Institute and Dept. of Computer Science, Paderborn University, Germany); Takamasa Higuchi (Toyota InfoTechnology Center, Co., Ltd., Tokyo, Japan); Onur Altintas (Toyota InfoTechnology Center, USA, Inc., Mountain View, California, US); Falko Dressler (Heinz Nixdorf Institute and Dept. of Computer Science, Paderborn University, Germany)
Umberto Fugiglando (MIT Senseable City Laboratory); Paolo Santi (IIT-CNR, MIT Senseable City Laboratory); Kacem Abida (VW Group Electronics Research); Sebastiano Milardo (University of Palermo); Carlo Ratti (MIT Senseable City Laboratory)
Mert D. Pese, Arun Ganesan, Kang G. Shin (University of Michigan – Ann Arbor)
- Infrastructure Bandwidth Allocation for Social Welfare Maximization in Future Connected Autonomous Vehicular Networks (20 min)
- Parked Cars as Virtual Network Infrastructure: Enabling Stable V2I Access for Long-Lasting Data Flows (20 min)
- Managing massive firmware-over-the-air updates for connected cars in cellular networks (20 min)
“Teng Liu (Department of Electrical, Computer, and Systems Engineering, Rensselaer Polytechnic Institute); Alhussein A. Abouzeid (Department of Electrical, Computer, and Systems Engineering, Rensselaer Polytechnic Institute); A. Agung Julius (Department of Electrical, Computer, and Systems Engineering, Rensselaer Polytechnic Institute)”
Florian Hagenauer (Heinz Nixdorf Institute and Dept. of Computer Science, Paderborn University, Germany); Christoph Sommer (Heinz Nixdorf Institute and Dept. of Computer Science, Paderborn University, Germany); Takamasa Higuchi (Toyota InfoTechnology Center, Co., Ltd., Tokyo, Japan); Onur Altintas (Toyota InfoTechnology Center, USA, Inc., Mountain View, California, US); Falko Dressler (Heinz Nixdorf Institute and Dept. of Computer Science, Paderborn University, Germany)
Carlos E. Andrade (AT&T Labs); Simon D. Byers (AT&T Labs);Vijay Gopalakrishnan (AT&T Labs); Emir Halepovic (AT&T Labs);Milap Majmundar (AT&T Labs); David J. Poole (AT&T Labs); Lien K. Tran (AT&T Labs);Christopher T. Volinsky (AT&T Labs)
- Poster: An In-Vehicle Software Defined Network Architecture for Connected and Automated Vehicles
- Poster: Receiver-Driven Semi-Broadcast for Vehicular Applications
- Poster: Investigating Doppler Effects on Vehicle-to-Vehicle Communication: An Experimental Study
- Poster: RSSI-Based Pedestrian Localization Using Artificial Neural Networks
- Demo: Radio Rate Transformer – A Portable VHF/UHF and Gigahertz Radio Combo Providing Broadband Ad-hoc Networking Services to Mobile Vehicles
- Demo: Freeway Merge Assistance System using DSRC
Peter Fussey (University of Sussex);George Parisis (University of Sussex)
PDohyung Kim (Sungkyunkwan University); Tae-Jin Lee (Sungkyunkwan University); Ikjun Yeom (Sungkyunkwan University)
Dwayne Jordan (East Tennessee State University); Nicholas Kyte (East Tennessee State University);Scott Murray (East Tennessee State University); Mohammad A Hoque (East Tennessee State University); Md Salman Ahmed (East Tennessee State University); Asad Khattak (University of Tennessee, Knoxville)
Mehdi Golestanian (University of Notre Dame);Christian Poellabauer (University of Notre Dame); Nitesh Chawla (University of Notre Dame)
Jiejun Kong (Turing Network Test L.L.C.)
Md Salman Ahmed (East Tennessee State University); Mohammad A Hoque (East Tennessee State University);Jackeline Rios-Torres (Oak Ridge National Lab); Asad Khattak (University of Tennessee, Knoxville)
- Reducing Unnecessary Pedestrian-to-Vehicle Transmissions Using A Contextual Policy (20 min)
- New Dimensions of Intersection Control with Connected and Automated Vehicles (invited paper, 15 min)
- Vulnerable Road User Protection through Intuitive Visual Cue on Smartphones (15 min)
Ali Rostami (WINLAB, Rutgers University); Bin Cheng (WINLAB, Rutgers University); Hongsheng Lu (Toyota InfoTech Center); Marco Gruteser (WINLAB, Rutgers University); John B. Kenney (Toyota InfoTech Center)
Yiheng Feng, Weili Sun, Jianfeng Zheng, and Henry X. Liu (University of Michigan, Ann Arbor)
Taeho Kim (Korea University); Wongoo Han (Korea University); Hyogon Kim (Korea University); Yongtae Park (Korea University)