Work

CoPreMo: A Collaborative Predictive Model in Time Series and Its Application to Radar Target Tracking for ADAS/AD Vehicles

Year: 2024

Type: article

Abstract: Automotive driving assistant systems (ADAS) and Automated driving (AD) technologies are widely used in unmanned vehicle control along a trajectory, and just like many other technologies, they present some risks such as the risk of misdirection and collision of the vehicle with an obstacle along its trajectory. To avoid these, many tracking systems such as the radar tracking system are used to detect and track targets and trajectories of ADAS/AD vehicles. However, the performance of the tracking operation is an important factor in determining the reliability of the tracking systems and ADAS/AD technologies, especially at the edge computing level. In this study, we proposed a tracking technique using a collaborative predictive model in time series, called CoPreMo, to improve the reliability of the tracking operation of a radar system. We carried out three experiments with the model on a simulated radar system to track the range of a target with different speeds in three ADAS/AD scenarios and achieve a range tracking error of 0.21m, 0.26m, and 0.32m, better than those of the baseline models.

Source: IEEE Transactions on Intelligent Vehicles

DOI: 10.1109/tiv.2024.3461730

Authors: Zie Eya Ekolle, Ryuji Kohno

Language: English

Topic: Radar Systems and Signal Processing

Subfield: Aerospace Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,495

Funder: National Institute of Information and Communications Technology