Vehicle Tracking in Outdoor Environment Based on Curvelet Domain

Object tracking from LiDAR point clouds, which are always incomplete, sparse, and unstructured, plays a crucial role in urban navigation. Some existing methods utilize a learned similarity network for locating the target, immensely limiting the advancements in tracking accuracy. In this study, we leveraged a powerful target discriminator and an accurate state estimator to robustly track target objects in challenging point cloud scenarios. Considering the complex nature of estimating the state, we extended the traditional Lucas and Kanade (LK) algorithm to 3D point cloud tracking. Specifically, we propose a state estimation subnetwork that aims to learn the incremental warp for updating the coarse target state. Moreover, to obtain a coarse state, we present a simple yet efficient discrimination subnetwork. It can project 3D shapes into a more discriminatory latent space by integrating the global feature into each point-wise feature. Experiments on KITTI and PandaSet datasets showed that compared with the most advanced of other methods, our proposed method can achieve significant improvements—in particular, up to 13.68% on KITTI.

Download Full-text

Evaluation of the heat changes in an ancient church because of restoration works: A microclimatic study supported by thermal images

Indoor and Built Environment ◽

10.1177/1420326x21994625 ◽

2021 ◽

pp. 1420326X2199462

Author(s):

Stefano Ridolfi ◽

Susanna Crescenzi ◽

Fabiana Zeli ◽

Stefano Perilli ◽

Stefano Sfarra

Keyword(s):

Methodological Approach ◽

Thermal Capacity ◽

Outdoor Environment ◽

Heat Index ◽

Heat Sources ◽

Thermal Images ◽

Roof Structure ◽

Period 2 ◽

Microclimate Monitoring ◽

The Church

This work is centred on an ancient Italian church. Since 2011, a restoration plan has been undertaken by following sequential phases. The methodological approach to restoration was guided by environmental monitoring campaigns. In particular, two thermo-hygrometric campaigns were carried out during the warm months of the years 2015 and 2016. The first set of measurements was executed during the restoration of facades and roofs, making it possible to reach even areas that are usually difficult to access. The second set was performed to evaluate the indoor thermo-hygrometric conditions following the work of the previous year. This was intended to assess their differences in variability, the influence of the outdoor environment and any real and perceived improvement. Results demonstrate that thermal images helped in identifying both the heat sources causing thermal discomforts and the good thermal capacity of masonries. Concerning the heat index (HI), the church showed an improvement in the trend of malaise perceived by people during the second summer period (∼2°C lower than 2015). Finally, in the last microclimate monitoring, the roof structure no longer acted as an amplifier for daily temperature excursions.

Download Full-text

Robust Quadrotor Control through Reinforcement Learning with Disturbance Compensation

Applied Sciences ◽

10.3390/app11073257 ◽

2021 ◽

Vol 11 (7) ◽

pp. 3257

Author(s):

Chen-Huan Pi ◽

Wei-Yuan Ye ◽

Stone Cheng

Keyword(s):

Neural Network ◽

Reinforcement Learning ◽

Control Strategy ◽

External Disturbance ◽

Control Agent ◽

Network Control ◽

Outdoor Environment ◽

Disturbance Compensation ◽

Tracking Accuracy ◽

Control Scheme

In this paper, a novel control strategy is presented for reinforcement learning with disturbance compensation to solve the problem of quadrotor positioning under external disturbance. The proposed control scheme applies a trained neural-network-based reinforcement learning agent to control the quadrotor, and its output is directly mapped to four actuators in an end-to-end manner. The proposed control scheme constructs a disturbance observer to estimate the external forces exerted on the three axes of the quadrotor, such as wind gusts in an outdoor environment. By introducing an interference compensator into the neural network control agent, the tracking accuracy and robustness were significantly increased in indoor and outdoor experiments. The experimental results indicate that the proposed control strategy is highly robust to external disturbances. In the experiments, compensation improved control accuracy and reduced positioning error by 75%. To the best of our knowledge, this study is the first to achieve quadrotor positioning control through low-level reinforcement learning by using a global positioning system in an outdoor environment.

Download Full-text

Hybrid PDA/FIR Filtering for Preceding Vehicle Tracking Using Automotive Radars

IEEE Access ◽

10.1109/access.2021.3107464 ◽

2021 ◽

pp. 1-1

Author(s):

Jung Min Pak

Keyword(s):

Vehicle Tracking ◽

Fir Filtering ◽

Preceding Vehicle ◽

Automotive Radars

Download Full-text

An Analysis of the External Validity of EEG Spectral Power in an Uncontrolled Outdoor Environment during Default and Complex Neurocognitive States

Brain Sciences ◽

10.3390/brainsci11030330 ◽

2021 ◽

Vol 11 (3) ◽

pp. 330

Author(s):

Dalton J. Edwards ◽

Logan T. Trujillo

Keyword(s):

External Validity ◽

Spectral Power ◽

Brain Activity ◽

Outdoor Environment ◽

Oscillatory Activity ◽

Brain State ◽

Beta Band ◽

Eeg Spectral Power ◽

Eeg Power ◽

Human Participants

Traditionally, quantitative electroencephalography (QEEG) studies collect data within controlled laboratory environments that limit the external validity of scientific conclusions. To probe these validity limits, we used a mobile EEG system to record electrophysiological signals from human participants while they were located within a controlled laboratory environment and an uncontrolled outdoor environment exhibiting several moderate background influences. Participants performed two tasks during these recordings, one engaging brain activity related to several complex cognitive functions (number sense, attention, memory, executive function) and the other engaging two default brain states. We computed EEG spectral power over three frequency bands (theta: 4–7 Hz, alpha: 8–13 Hz, low beta: 14–20 Hz) where EEG oscillatory activity is known to correlate with the neurocognitive states engaged by these tasks. Null hypothesis significance testing yielded significant EEG power effects typical of the neurocognitive states engaged by each task, but only a beta-band power difference between the two background recording environments during the default brain state. Bayesian analysis showed that the remaining environment null effects were unlikely to reflect measurement insensitivities. This overall pattern of results supports the external validity of laboratory EEG power findings for complex and default neurocognitive states engaged within moderately uncontrolled environments.

Download Full-text

Application Method of a Simplified Heat and Moisture Transfer Model of Building Construction in Residential Buildings

Energies ◽

10.3390/en14144180 ◽

2021 ◽

Vol 14 (14) ◽

pp. 4180

Author(s):

Joowook Kim ◽

Michael Brandemuehl

Keyword(s):

Latent Heat ◽

Moisture Transfer ◽

Residential Buildings ◽

Building Energy ◽

The United States ◽

Building Envelope ◽

Outdoor Environment ◽

Transfer Model ◽

Heat And Moisture Transfer ◽

Heat And Moisture

Several building energy simulation programs have been developed to evaluate the indoor conditions and energy performance of buildings. As a fundamental component of heating, ventilating, and air conditioning loads, each building energy modeling tool calculates the heat and moisture exchange among the outdoor environment, building envelope, and indoor environments. This paper presents a simplified heat and moisture transfer model of the building envelope, and case studies for building performance obtained by different heat and moisture transfer models are conducted to investigate the contribution of the proposed steady-state moisture flux (SSMF) method. For the analysis, three representative humid locations in the United States are considered: Miami, Atlanta, and Chicago. The results show that the SSMF model effectively complements the latent heat transfer calculation in conduction transfer function (CTF) and effective moisture penetration depth (EMPD) models during the cooling season. In addition, it is found that the ceiling part of a building largely constitutes the latent heat generated by the SSMF model.

Download Full-text