Target Tracking based on Millimeter Wave Radar in Complex Scenes

Abstract

Currently, the method of using millimeter wave radar to detect obstacles in front of vehicles has been widely used. When using millimeter wave radar to detect obstacles on the road, the radar has more noise interference due to the changeable road environment and complex background. Combined with the complexity and variety of road targets, the random changes of scattering intensity and relative phase of different parts cause the distortion of the echo phase wave, resulting in the flicker noise that affects the accuracy of measurement, and even lead to the loss of targets. In this case, there are some shortcomings in tracking the target using the ordinary Kalman filter algorithm. In this paper, a Sage-Husa adaptive Kalman filtering algorithm is designed for the road environment to track radar targets and improve the accuracy of target tracking. Then, the radar and machine vision information fusion method is used to intuitively judge the filtering effect and determine whether the radar loses the target. Finally, the true value of the target position is approximated by filtered value when the radar loses its target. The experimental results show that this method can improve the accuracy and reliability of the millimeter wave radar.

Millimeter Wave Sensing: A Review of Application Pipelines and Building Blocks

“…Several Bayesian filters exist, such as the custom Bayesian [45], [185], particle [44], [136], α − β [128], Kalman [46], [53], [58], extended Kalman [55], [94], [96], [114], [115], [135], [186], fusion extended Kalman [95], [146], unscented Kalman [89], fusion adaptive Kalman [51], and adaptive Sage-Husa Kalman [52], [61] filter. The Kalman filter [219], under linear, quadratic, and Gaussian assumptions, can represent the state transition and observation functions x t = g(x t−1 , pn t ) and s t = h(x t , mn t ) as a set of linear equations.…”

“…Some pipelines apply sensor fusion of millimeter wave data with data originating from other sensing domains. Sensing domains include vision [51], [61], [66], [67], [71], [95], [105], [120], [146], depth [44], [105], lidar [71], inertial measurements [89], and exerted forces [91]. Data originating from these domains complement millimeter wave data and therefore cause more accurate analytical modeling performance in several situations.…”

“…Data originating from these domains complement millimeter wave data and therefore cause more accurate analytical modeling performance in several situations. Analytical modeling performance in on-road detection and tracking suffers from limited spatial resolution of, and noise in, millimeter wave data [51], [61], [67], [95], [120]. Odometry information model performance also suffers from noise in millimeter wave data [89].…”

Millimeter Wave Sensing: A Review of Application Pipelines and Building Blocks

“…Several Bayesian filters exist, such as the custom Bayesian [45], [185], particle [44], [136], α − β [128], Kalman [46], [53], [58], extended Kalman [55], [94], [96], [114], [115], [135], [186], fusion extended Kalman [95], [146], unscented Kalman [89], fusion adaptive Kalman [51] and adaptive Sage-Husa Kalman [52], [61] filter. The Kalman filter [215], under linear, quadratic and Gaussian assumptions, can represent the state transition and observation functions x t = g(x t−1 , pn t ) and s t = h(x t , mn t ) as a set of linear equations.…”

“…Some pipelines apply sensor fusion of millimeter wave data with data originating from other sensing domains. Sensing domains include vision [51], [61], [66], [67], [71], [95], [105], [120], [146], depth [44], [105], lidar [71], inertial measurements [89] and exerted forces [91]. Data originating from these domains complement millimeter wave data and therefore cause more accurate analytical modeling performance in several situations.…”