chrxh/alien
Alien
Evolve is an evolution-based organism designer and GPU-accelerated artificial life simulator that combines interactive particle physics with a real-time simulation editor. At its core, it runs genetic algorithm evolution on self-replicating graph structures to evolve digital organisms, offloading particle physics, neural networks, and rendering entirely to the GPU through a compute shader pipeline for real-time performance.
The project distinguishes itself with graph-based organism design that uses a directed graph editor to visually define organism structure, connections, and neural controllers, all manipulable live during simulation. Spatiotemporal parameter variation systematically changes simulation parameters across spatial regions and over time to create heterogeneous environments, while headless simulation execution allows running evolution from the command line with configurable parameters and time steps.
Beyond its core identity, Evolve provides particle physics simulation handling soft body dynamics, fluids, heat, and damage, along with neural network particle control that enables learned behavior without hard-coded rules. It includes interactive editors for designing agents, environments, and custom organisms, plus a built-in community simulation browser for discovering, sharing, and upvoting simulations.
Features
- Digital Organism Designers - Evolves multi-cellular digital organisms through mutation and natural selection within a shared environment.
- Agent Simulation Environments - Provides a built-in graph and genome editor for designing custom artificial life agents and their environments.
- Custom Organism Designers - Ships a graph and genome editor for visually constructing custom digital organisms with user-defined behaviors.
- Genetic Algorithms - Runs mutation and selection on self-replicating graph structures to evolve digital organisms over generations.
- Multi-Cellular Organism Modeling - Creates multi-cellular organisms with genetic inheritance, cell-by-cell construction, and neural network control.
- Particle Neural Network Controllers - Assigns artificial neural networks to particles, enabling learned behavior without hard-coded rules.
- Self-Replicating Structure Mutation - Subjects self-replicating structures to random mutations in a shared environment, enabling natural selection.
- Organism Graph Editors - Uses a directed graph editor to visually define organism structure, connections, and neural controllers.
- Interactive Simulations - Provides a real-time interactive editor for manipulating particles, connections, and organism blueprints during simulation.
- Compute Shader Pipelines - Executes particle physics, neural networks, and rendering entirely through GPU compute and fragment shaders.
- Particle Physics Simulations - Simulates soft body physics, fluids, and particle interactions in a dynamic environment with damage and gluing mechanics.
- GPU-Accelerated Physics Simulations - Provides GPU-accelerated particle physics and fluid simulation for soft body dynamics, heat, and damage in real-time.
- Artificial Life Simulators - A platform for simulating evolving digital organisms and particle physics using GPU compute for real-time performance.
- Ecosystem Particle Simulators - Uses GPU hardware to handle millions of interacting particles in real time for ecosystem simulation.
- Interactive Physics Engines - Simulates soft and rigid body dynamics, fluids, heat, and damage using GPU acceleration for large-scale ecosystems.
- Digital Life Evolution - Evolves digital organisms with neural networks and genetic algorithms to develop complex behaviors without manual programming.
- Spatiotemporal Parameter Variation - Systematically changes simulation parameters across different spatial regions and over time to create heterogeneous environments.
- Spatiotemporal Parameter Variations - Implements spatiotemporal parameter variation to dynamically change simulation properties across different regions and time steps.
- Simulation Editors - Provides graph and genome editors to design custom agents, environments, and modify simulations interactively.
- Organism Design Interfaces - Designs custom agents and environments visually by editing genomes, connection graphs, and simulation parameters in real-time.
- Headless Execution - Runs simulations from the command line with configurable parameters, input files, and time steps.
