Reliable Heading Tracking for Pedestrian Road Crossing Prediction using Commodity Devices > 자유게시판

• 목록
• 답변
• 글쓰기
• 게시판 리스트 옵션
  • 수정
  • 삭제

Pedestrian heading tracking permits applications in pedestrian navigation, traffic security, and accessibility. Previous works, utilizing inertial sensor fusion or machine learning, are restricted in that they assume the cellphone is fixed in specific orientations, hindering their generalizability. We suggest a new heading tracking algorithm, the Orientation-Heading Alignment (OHA), iTagPro smart device which leverages a key perception: individuals have a tendency to hold smartphones in sure ways as a result of habits, similar to swinging them whereas walking. For each smartphone perspective throughout this motion, OHA maps the smartphone orientation to the pedestrian heading and learns such mappings efficiently from coarse headings and smartphone orientations. To anchor itagpro locator our algorithm in a sensible state of affairs, we apply OHA to a challenging process: predicting when pedestrians are about to cross the road to improve street user security. Particularly, utilizing 755 hours of walking information collected since 2020 from 60 individuals, we develop a lightweight mannequin that operates in actual-time on commodity devices to predict street crossings. Our evaluation exhibits that OHA achieves 3.Four occasions smaller heading errors across 9 scenarios than existing strategies.

Furthermore, OHA allows the early and correct detection of pedestrian crossing behavior, issuing crossing alerts 0.35 seconds, iTagPro device on average, earlier than pedestrians enter the street vary. Tracking pedestrian heading involves continuously tracking an individual’s dealing with direction on a 2-D flat airplane, usually the horizontal aircraft of the global coordinate system (GCS). Zhou et al., 2014). For ItagPro instance, a pedestrian could possibly be walking from south to north on a highway whereas swinging a smartphone. On this case, smartphone orientation estimation would point out the device’s dynamic orientation relative to the GCS, commonly represented by Euler angles (roll, pitch, yaw). On the other hand, tracking pedestrian heading ought to accurately present that the pedestrian is moving from south to north, regardless of how the smartphone is oriented. Existing approaches to estimating pedestrian heading by IMU (Inertial Measurement Unit) make use of a two-stage pipeline: first, they estimate the horizontal aircraft using gravity or magnetic fields, and then integrate the gyroscope to trace relative heading changes (Manos et al., 2018; Thio et al., 2021; Deng et al., 2015). These approaches hinge on a essential assumption: the cellphone must stay static relative to the pedestrian body.

We propose a brand new heading monitoring algorithm, Orientation-Heading Alignment (OHA), which leverages a key perception: individuals have a tendency to carry smartphones in certain attitudes as a consequence of habits, whether or not swinging them while walking, stashing them in pockets, or inserting them in baggage. These attitudes or relative orientations, defined because the smartphone’s orientation relative to the human body rather than GCS, primarily depend on the user’s habits, iTagPro locator characteristics, or iTagPro locator even clothing. For instance, regardless of which course a pedestrian faces, they swing the smartphone in their habitual manner. For every smartphone attitude, OHA maps the smartphone orientation to the pedestrian heading. Because the attitudes are relatively stable for every particular person (e.g., holding a smartphone in the proper hand and swinging), it is possible to be taught the mappings efficiently from coarse headings and smartphone orientation. Previous research (Liu et al., iTagPro locator 2023; Yang et al., iTagPro locator 2020; Lee et al., 2023) has famous a similar perception but adopted a different approach for heading tracking: gathering IMU and correct heading info for a number of smartphone attitudes and coaching a machine learning model to predict the heading.

However, due to system discrepancies and varying consumer behaviors, it isn't feasible to construct a machine studying model that generalizes to all potential smartphone attitudes. To anchor our heading estimation algorithm in a sensible scenario, we apply OHA to a difficult process: predicting when pedestrians are about to cross the highway-an important downside for iTagPro technology bettering street consumer safety (T., pril; Zhang et al., 2021, 2020). This activity, which requires accurate and well timed predictions of pedestrian crossings, ItagPro is additional complicated by the numerous crossing patterns of pedestrians and the complexity of highway layouts. Based on the OHA heading, we suggest PedHat, iTagPro locator a lightweight, infrastructure-free system that predicts when a pedestrian is about to cross the closest highway and points crossing alerts. PedHat incorporates a lightweight mannequin that accepts OHA headings as inputs and operates in real-time on consumer devices to predict road crossings. We developed this mannequin using data we collected since 2020 from 60 individuals, every contributing two months of traces, masking 755 hours of walking data.