Procedural Video Game Scene Generation by Genetic and Neutrosophic WASPAS Algorithms

The demand for automated game development assistance tools can be fulfilled by computational creativity algorithms. The procedural generation is one of the topics for creative content development. The main procedural generation challenge for game level layout is how to create a diverse set of levels that could match a human-crafted game scene. Our game scene layouts are created randomly and then sculpted using a genetic algorithm. To address the issue of fitness calculation with conflicting criteria, we use weighted aggregated sum product assessment (WASPAS) in a single-valued neutrosophic set environment (SVNS) that models the indeterminacy with truth, intermediacy, and falsehood memberships. Results are presented as an encoded game object grid where each game object type has a specific function. The algorithm creates a diverse set of game scene layouts by combining game rules validation and aesthetic principles. It successfully creates functional aesthetic patterns without specifically defining the shapes of the combination of games’ objects.

AN IMPLEMENTATION OF THE PROCEDURAL GENERATION OF UNSTRUCTURED QUADRANGULAR GRID ALGORITHM BASED ON A STRUCTURED TRIANGULAR GRID IN REGULAR HEXAGONS ON THE PLANE

Работа посвящена реализации алгоритма процедурной генерации нерегулярной четырехугольной сетки, позволяющего рассчитывать сетку для большой области в реальном времени. При генерации используются кубическая система координат, в которой строится регулярная треугольная сетка для каждой ячейки шестиугольной сетки, процедура релаксации четырехугольной сетки. This work is devoted to the implementation of an algorithm for procedural generation of an unstructured quadrangular grid, which allows to calculate the grid for a large area in real time. When building the grid, a cubic coordinate system, in which a structured triangular grid is built for each cell of a hexagonal grid, and a relaxation of the quadrangular grid algorithm are used.

Bits + Pieces : Investigating the Procedural Ideation of Architectural Massing for Early Stage Design

<p>The architectural discipline is constantly experiencing change to the way in which its practitioners operate. The continual evolution of computing hardware and the substantial development of Computer Aided Architectural Design (CAAD) has seen Architecture shift from a discipline of predominantly analogue techniques to one that relies almost entirely on the digital medium. As a result, the role of the practicing architect has seen considerable change. Architecture, once a discipline of pencil and paper, now shares creative techniques and tools with Computer Science, Film, Visual Effects, Interactive Media, Robotics, and Computer programming. Such new partners are providing alternative views of what it is to be a creative practitioner, challenging the discipline of architecture to step beyond the preconceived boundaries and means of operating embodied within conventional practice. Architects now have the opportunity to adopt new methods for the production of the built environment. This research engages with developing computational techniques designed for film and interactive media and explores how they can be utilised to augment the way in which architecture may be produced. This body of researches adopts the technique procedural generation as a vehicle for this investigation; a technique used for content creation in interactive media and game design. This research also adopts the use of a computational design software called Houdini - an industry standard procedural software used widely within film and game. Through an architectural lens, it explores the re-purposing of this software and procedural design, developing an understanding for how they can both aid in the ideation of built form during the infancy of the design process. This research initially addressed the question: ‘how can conventional architectural practices be augmented by procedural computational design techniques, to further explore the impacts of opportunity and ideation on architectural design?’ As a result of refinement, it came around to focus on asking ‘how can the application of procedural generation design techniques augment the ideation of architectural massing for early stage design?’ It identifies how procedural techniques can be used in the process of ideating architecture and aims to investigate how procedural generation offers an alternative methodology to the production of architecture in early design stages. It explores, through computational design, the limitations and constraints that occur in the process of mastering design orientated procedural techniques. It subsequently develops, through computational design, an understanding of how procedural techniques can be applied to the early stage design of architecture. Finally, through architectural design, it examines how procedural design techniques can be partnered with specific architectural conditions such as site, function, and form, in order to augment the architectural ideation process.</p>

Bits + Pieces : Investigating the Procedural Ideation of Architectural Massing for Early Stage Design

<p>The architectural discipline is constantly experiencing change to the way in which its practitioners operate. The continual evolution of computing hardware and the substantial development of Computer Aided Architectural Design (CAAD) has seen Architecture shift from a discipline of predominantly analogue techniques to one that relies almost entirely on the digital medium. As a result, the role of the practicing architect has seen considerable change. Architecture, once a discipline of pencil and paper, now shares creative techniques and tools with Computer Science, Film, Visual Effects, Interactive Media, Robotics, and Computer programming. Such new partners are providing alternative views of what it is to be a creative practitioner, challenging the discipline of architecture to step beyond the preconceived boundaries and means of operating embodied within conventional practice. Architects now have the opportunity to adopt new methods for the production of the built environment. This research engages with developing computational techniques designed for film and interactive media and explores how they can be utilised to augment the way in which architecture may be produced. This body of researches adopts the technique procedural generation as a vehicle for this investigation; a technique used for content creation in interactive media and game design. This research also adopts the use of a computational design software called Houdini - an industry standard procedural software used widely within film and game. Through an architectural lens, it explores the re-purposing of this software and procedural design, developing an understanding for how they can both aid in the ideation of built form during the infancy of the design process. This research initially addressed the question: ‘how can conventional architectural practices be augmented by procedural computational design techniques, to further explore the impacts of opportunity and ideation on architectural design?’ As a result of refinement, it came around to focus on asking ‘how can the application of procedural generation design techniques augment the ideation of architectural massing for early stage design?’ It identifies how procedural techniques can be used in the process of ideating architecture and aims to investigate how procedural generation offers an alternative methodology to the production of architecture in early design stages. It explores, through computational design, the limitations and constraints that occur in the process of mastering design orientated procedural techniques. It subsequently develops, through computational design, an understanding of how procedural techniques can be applied to the early stage design of architecture. Finally, through architectural design, it examines how procedural design techniques can be partnered with specific architectural conditions such as site, function, and form, in order to augment the architectural ideation process.</p>

MarsExplorer: Exploration of Unknown Terrains via Deep Reinforcement Learning and Procedurally Generated Environments

This paper is an initial endeavor to bridge the gap between powerful Deep Reinforcement Learning methodologies and the problem of exploration/coverage of unknown terrains. Within this scope, MarsExplorer, an openai-gym compatible environment tailored to exploration/coverage of unknown areas, is presented. MarsExplorer translates the original robotics problem into a Reinforcement Learning setup that various off-the-shelf algorithms can tackle. Any learned policy can be straightforwardly applied to a robotic platform without an elaborate simulation model of the robot’s dynamics to apply a different learning/adaptation phase. One of its core features is the controllable multi-dimensional procedural generation of terrains, which is the key for producing policies with strong generalization capabilities. Four different state-of-the-art RL algorithms (A3C, PPO, Rainbow, and SAC) are trained on the MarsExplorer environment, and a proper evaluation of their results compared to the average human-level performance is reported. In the follow-up experimental analysis, the effect of the multi-dimensional difficulty setting on the learning capabilities of the best-performing algorithm (PPO) is analyzed. A milestone result is the generation of an exploration policy that follows the Hilbert curve without providing this information to the environment or rewarding directly or indirectly Hilbert-curve-like trajectories. The experimental analysis is concluded by evaluating PPO learned policy algorithm side-by-side with frontier-based exploration strategies. A study on the performance curves revealed that PPO-based policy was capable of performing adaptive-to-the-unknown-terrain sweeping without leaving expensive-to-revisit areas uncovered, underlying the capability of RL-based methodologies to tackle exploration tasks efficiently.

PROCEDURAL GENERATION OF MASSIVE 3D GEOMETRY USING IMPROVED MARCHING CUBES ALGORITHM

Процедурная генерация, или создание контента во время работы программы, это сложное направление, которое требует не только понимания 3D-графики, но и навыков программирования графики, что часто сводится к изучению работы графических процессоров. Из-за такой сложности разработчики часто используют уже готовые инструменты для создания контента. Такие инструменты обобщают и упрощают работу, предоставляя большой заготовленный набор функции, который можно использовать не зная программирования вовсе. К сожалению, обобщение часто приводит к уменьшению гибкости и вводит новые ограничения. Статистика показывает, что использование процедурной генерации, для создания массивной 3D-геометрии, невозможно при использовании готовых инструментов с уже заготовленными функциями. Такие инструменты не позволяют воплотить огромные масштабы массивной геометрии в жизнь из-за различных ограничений. Кроме того, существующие алгоритмы создания 3D-геометрии часто не учитывают применение этих алгоритмов для создания массивной 3D-геометрии, например, планет. Рассматриваемый в этой работе алгоритм Marching Cubes также не учитывает применение алгоритма для создания массивной геометрии, из-за чего применение этого алгоритма в таких целях будет иметь много ограничений и много недостатков. Но данный алгоритм выбран не случайно, он обладает большой популярностью и мы поговорим почему. Данная работа фокусируется на представлении новой модификации на уже существующий алгоритм Marching Cubes в целях применения его в рамках массивной геометрии. Данный алгоритм найдет применение в компьютерных играх с космической тематикой, наш алгоритм позволяет создавать массивную 3D-геометрию планетарных масштабов даже на слабых компьютерах без особых затрат по ресурсам. Кроме того, наш алгоритм позволяет изменять сгенерированную геометрию в реальном времени, без задержек по времени, что так важно компьютерным играм. Procedural generation, or the creation of content while a program is running, is a complex area that requires not only an understanding of 3D graphics, but also graphics programming skills, which often boils down to learning how GPUs work. Because of this complexity, developers often use off-the-shelf content creation tools. Such tools generalize and simplify work by providing a large pre-built set of functions that can be used without knowing programming at all. Unfortunately, generalization often reduces flexibility and introduces new constraints. Statistics show that using procedural generation to create massive 3D geometry is impossible when using ready-made tools with already prepared functions. Such tools do not allow the huge scales of massive geometry to be brought to life due to various constraints. In addition, existing 3D geometry creation algorithms often do not account for the application of these algorithms to create massive 3D geometry such as planets. The Marching Cubes algorithm considered in this work also does not take into account the use of the algorithm for creating massive geometry, which is why the use of this algorithm for such purposes will have many limitations and many disadvantages. But this algorithm was not chosen by chance, it is very popular and we will talk why. This work focuses on modifying the existing Marching Cubes algorithm to apply it to massive geometry. This algorithm will find application in computer games with a space theme, our algorithm allows to create massive 3D geometry of planetary scales even on a low-end computers without special resource costs. In addition, our algorithm allows to change the generated geometry in real time, without time delays, which is so important for computer games.