MFI 2022 - Special Sessions

If you are interested in presenting your paper in a special session please email info@mfi2022.com 

SpSe 1: Sensor Fusion for Autonomous Vehicles

 

Proposers:

Augie Widyotriatmo, Institut Teknologi Bandung.

Agus Hasan, Norwegian Norwegian University of Science and Technology.

Vadim Kramar, Oulu University of Applied Science.

Description:

Multi-sensor fusion and integration have shown many advantages over single sensor in various applications, including in autonomous vehicles. For autonomous vehicles (e.g., autonomous cars, surface vehicles, drones), multi-sensor fusion and integration can be applied for localization, mapping, perception recognition, control and decision making, etc. In localization system, the sensor fusion technique integrates global coordinate sensors, such as global navigation satellite system (GNSS), high-definition map, or LIDAR-based navigation sytem, and inertial-based sensors, such as encoders or inertial measurement unit (IMU), to improve the accuracy and sampling rate of measurement. The fusion between LIDAR and camera can be used to improve the perception of autonomous vehicles. The decision-making and control systems need to combine several sensors to recognize the environment, namely sensors that have close-range characteristics with multi-angle rays, sensors with long-range and single-beam characteristic, as well as a system recognition system using deep-learning algorithm. This session welcomes researchers that would like to contribute in the advancement of multi-sensor fusion and integration in the applications of autonomous vehicles including, but not limited to, the integration of information from multi-sensors, new technique in sensor fusion algorithm, the use of multi-sensors fusion and integration for control, decision making, and fault detection, in autonomous vehicles.

SpSe 2: Industrial Applications

Proposers:

Dr. Ramesh Bharadwaj, US Naval Research Laboratory.

Prof. Ram Narayanan, Pennsylvania State University.

Description:

Machine Learning methodologies have produced more accurate detection, including a significant reduction in false positive rates, through data fusion. Upstream Data Fusion (UDF) methods utilize artificial intelligence shown to consistently improve accuracy in women’s health applications. Data from Digital Breast Tomosynthesis (DBT), fused with ultrasound data, gave the best results. The objective of this special session is to initiate discussion on the industrial

application of ML based fusion methods and their underlying theorerical foundation.