4466 research outputs found
Sort by
A Reinforcement Learning Framework for Motion Planning of Autonomous Vehicles
No full textThe paper introduces a framework that has been developed for
the design and verification of motion planning algorithms for autonomous
driving. The framework allows for the use of reinforcement learning for autonomous
driving that requires complex and computationally intensive simulations.
The key element in the presented approach plays a multi-agent
closed-loop simulation of the traffic environment. Using the framework, the
training process can be performed in parallel on high-performance computing
clusters. Therefore, the framework provides an easy way to explore the
potential of reinforcement learning for autonomous driving applications
A Study on Color Application and Perception in Urban Neighborhood Renewal Based on Simulation Comparison: Taking the Case of Kuangdaxuefu Street in Xuzhou, China
No full textFor the emotional needs of architectural color diversity in the renewal of urban micro-scale areas in China, this study intends to explore the trends in color perception which are influenced by changes in architectural color characteristics. In this study, the main color characteristics that influence the urban color renewal were determined through the preliminary basic investigation in China. The color psychological perception experiments were conducted by combining VR technology with questionnaire, so as to determine the correlation between architectural color characteristics and main color perceptions, as well as the correlation between different color perceptions
Improvement of Attention Mechanism Explainability in Prediction of Chemical Molecules’ Properties
No full textIn this paper, the analysis of selected graph neural network operators
is presented. The classic Graph Convolutional Network (GCN) was
compared with methods containing trainable attention coefficients: Graph
Attention Network (GAT) and Graph Transformer (GT). Moreover, which
is an original contribution of this work, training of GT was modified with
an additional loss function component enabling easier explainability of the
produced model. The experiments were conducted using datasets with chemical
molecules where both classification and regression tasks are considered.
The results show that additional constraint not only does not make the results
worse but, in some cases, it improves predictions
Recognition of Shoplifting Activities in CCTV Footage Using the Combined CNN-RNN Model
No full textThe recognition of human activities through surveillance has numerous
applications across various fields. This article presents a proposed
approach to identify shoplifting in camera-recorded video data using a neural
classifier that combines two neural networks, specifically, convolutional
and recurrent networks. The hybrid architecture consists of two parallel
streams: initial and processed video fragments (histogram of oriented gradients
and optical flow). The convolutional network extracts features from
each frame of the video fragment, while the recurrent network processes the
temporal information from sequences of frames as features to classify the
activity
Analysis of Surface EMG Signals to Control of a Bionic Hand Prototype
No full textThe aim of the presented project is to develop a comprehensive
system for acquiring surface EMG data and carry out time-frequency analysis
to determine useful parameters for subsequent gesture classification for
a simple bionic hand prosthesis. This system is expected to assist in controlling
both the prosthetic hand and the robotic hand in making precise
gestures with the fingers on the hand. The article presents the methods for
acquiring and processing multi-channel EMG signals and feature extraction
for gesture recognition by an artificial neural network (ANN)
The collective construction of planning in small cities– the experience of technical assistance in the construction of the Master Plan for Carnauba
No full textThe main instrument of municipal urban planning in Brazil is the Master Plans. They provide the guidelines for the city’s urban growth. Every municipality with more than 20 thousand inhabitants, with public interest, touristic and environmental potential or with environmentally degraded areas must have a PD1, which, by law, must be elaborated in a participatory way involving all segments of the population through public hearings and community meetings. One important critical and transforming agent of urban space is the University, fulfilling its social role as a producer of knowledge. This article presents partial results of studies carried out in a city in the interior of Ceará, called Carnaubal, with a population estimated at 17,763 inhabitants and located 345.8 km from the state’s capital, Fortaleza.
Carnaubal is located in the administrative region of Ibiapaba and, although it does not require the elaboration of a Master Plan, it is extremely important to structure the growth of these small towns in the interior of the country from now on, in order to minimize the harmful effects of their disorderly growth on the environment. In addition, the instruments present in the Master Plan enable the optimization of public investment both in terms of urban design and in the urban economy and distribution of public services. The proposal for technical assistance in the elaboration of Carnaubal’s Master Plan arose at the request of a former student and counselor who showed interest and concern with the planning of her city. From this initiative, the research group ‘Urbanismo Colaborativo – UrbColab’, affiliated with the research center ‘Núcleo de Pesquisa em Arquitetura, Cidade e Tecnologia – NuPACT’/CNPq2, began this experience involving students of Architecture and Urbanism, teachers, professionals and the population, using various methods, drone assistance, Qgis3, and fieldwork
Socio-cognitive Flock-based Optimization
No full textA novel optimization algorithm inspired by socio-cognitive phenomena
and based on flock architecture is presented along with promising
preliminary experimental results
Urban Morphology and Anthropology – Synergies and a Joint Language
No full textUrban anthropology is a sub-branch of social anthropology, and works, like urban morphology, with the city as field of study. While urban morphology is concerned with physical form, urban elements and pattern analyses over time, anthropology’s perspective is that of the social and cultural in the present moment. Anthropology investigates people’s experiences and meaning-making, and can bring a humanistic and qualitative lens to historical morphological studies, as well as futuristic planning and urban design. This paper looks at literature and theories in social anthropology and urban morphology to discuss and encourage convergence in morphological, anthropological, planning and urban design discourses. Pierre Bourdieu's theories are widely used by anthropologists and his concepts of habitus and social space are relevant to morphology. Habitus are the embodied dispositions that guide individuals in their behaviours and has been described as a sense of one’s own and other’s place. Social space is the physical space where relations between different habit uses play out. These two concepts are of interest to the study of spatiality, emplacement and mobility and can inform how we think about physical and social environments in relation to each other. This paper aims to find a joint language and create synergy between urban morphology and urban anthropology and contribute with holistic and humanistic approaches in analysing cities, city planning and urban design