2020 International Conference On Computer Aided Design

The Premier Conference Devoted to Technical Innovations in Electronic Design Automation

November 2-5, 2020VIRTUAL CONFERENCE

MP Associates, Inc.
WEDNESDAY November 04, 9:00am - 9:30am | Slot 3
EVENT TYPE: SPECIAL SESSION
SESSION 12C
Towards Real-time Energy-efficient Mobile Robotics: From Algorithm to Hardware
Moderator:
Jiang Hu - Texas A&M Univ.
Organizer:
Bo Yuan - Rutgers Univ., Piscataway, NJ
Background— A Robot is a very complex information processing system. A robot needs inception capability to sense, cognition capability to understand, and decision capability to act. From the perspective of computing, such an entire sense-understand-act pipeline is involved with various computation-intensive algorithms in many fields, including but not limited to machine learning, computer vision, control, signal processing etc. Motivation— In the emerging cyber-physical system (CPS) era, a rapidly growing field in robotics, namely mobile robotics, has received a lot of attention because of its very widespread practical applications, such as search-and-rescue, underwater exploration, and autonomous delivery. In the scenario of mobile robotics, computational resource and energy budget are typically restricted, while the need for real-time and low-power solutions become more demanding, thereby bringing severe challenges but also exciting opportunities for the VLSI and computing community. Although the study in efficient computing for robotics is still in its infancy, recently researchers in both academia and industry have made significant progress in addressing the above challenges. These research efforts that would be presented in this special session, range from algorithm to circuit to accelerator architecture to near-sensor computing, and they will lead to promising cross-layer solutions towards real-time energy-efficient mobile robotics. Rationale-- This special session will be a good match for ICCAD 2020. Mobile robotics is naturally involved with many computing limitations and challenges, where the VLSI and computing community can provide promising solutions. In addition, many classical problems in robotics, such as motion planning and grid-based path routing, share the very similar mathematical format in the problems well investigated in VLSI field, e.g. the routing problem in CAD. According to the initial feedbacks and the received topics of all the prospective contributors, we are confident that this special session we hereby propose will be of significant interest and importance to ICCAD 2020 attendees and be of high quality and impact commensurate with the high reputation of ICCAD. Moreover, we also hold the belief that this special session is very necessary. The research efforts on energy-efficient computing for mobile robotics are quite interdisciplinary, which requires a comprehensive investigation ranging from algorithm innovation to underlying hardware circuit design. To sum up, the wide spectrum of those research efforts makes us believe that it is necessary to organize this special session in ICCAD 2020.

12C.1Efficient Computing for Low-Energy Robotics
 Speaker: Vivienne Sze - MIT
 Author: Vivienne Sze - MIT
12C.2Hardware Acceleration of Robot Scene Perception Algorithms
 Speaker: R. Iris Bahar - Brown Univ., Providence, RI
 Authors: Yanqi Liu - Brown Univ.
Can Eren Durman - Brown Univ.
Giuseppe Calderoni - Politecnico di Torino
R. Iris Bahar - Brown Univ., Providence, RI
12C.3DaDu Series- Fast and Efficient Robot Accelerators
 Speaker: Yinhe Han - Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Scie
 Authors: Yinhe Han - Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Scie
Yuxin Yang - Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Scie
Xiaoming Chen - Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Scie
Shiqi Lian - Center for Intelligent Computing Systems, Institute of Computing Technology, Chinese Academy of Scie
12C.4TACOS: Tactile Core with Optical Strain Sensor for Robotic Smart Skin
 Speaker: Fei Qiao - Tsinghua Univ.
 Authors: Zheyu Liu - Tsinghua Univ., Beijing, China
Jingyi Zhou - Tsinghua Univ., Beijing, China
Huichan Zhao - Tsinghua Univ.
Qi Wei - Tsinghua Univ., Beijing, China
Fei Qiao - Tsinghua Univ.
Xinjun Liu - Tsinghua Univ., Beijing, China
Huazhong Yang - Tsinghua Univ., Beijing, China