Skip to main content
Skip to main content

PREHISTORICSIMULATEDINTELLIGENCE

Open-source platform for dinosaur-inspired locomotion research
powered by physics simulation and reinforcement learning

GET STARTEDVIEW ON GITHUB (opens in new tab)
SPECIES
3
ACTUATORS
73
PUBLISHED RUN STEPS
96M

Core Features

MuJoCo Physics

Physics-based articulated models inspired by dinosaur morphology, with contact dynamics and actuator models.

Reinforcement Learning

PPO and SAC algorithms via Stable-Baselines3 with automated 3-stage curriculum learning.

3 Species

Velociraptor (22 actuators), T-Rex (21 actuators), Brachiosaurus (30 actuators).

Open Source

MIT licensed with TOML configs, Gymnasium registration, W&B tracking, and Docker support.

How It Works

Each species learns through a 3-stage curriculum, progressively mastering more complex behaviors

STAGE 1

Balance

Stand & Stabilize

The agent learns to stand upright and maintain balance using proprioceptive feedback and joint torques.

STAGE 2

Locomotion

Walk & Run

Building on balance skills, the agent develops forward locomotion with coordinated movement patterns.

STAGE 3

Behavior

Species-Specific Task

Species-specific simulator tasks add sickle-claw contact, a fixed head-contact proxy, or head-tip target proximity.

See It In Action

PPO Stage 1 — Balance
Velociraptor learning to balance using PPO reinforcement learning
Historical, unverified PPO / Stable-Baselines3 artifact; backend version not recorded. Not evidence for the current model or configuration.

Published Run Summaries

Provenance-labelled metrics and the curriculum videos currently available in the repository

Sickle-claw contact attacks
Task success
93%
Max reported avg. velocity
3.47 m/s
Timesteps
22M
Run
PPO · Stable-Baselines3 · 2026-03-15

historical model · unverified · evaluation episode count not recorded · Stable-Baselines3 version not recorded. Metrics are retained experiment records, not claims about the current model revision.

STAGE 1PPO · Stable-Baselines3 · historical

Balance

Learn to stand and balance without falling

unverified artifact; backend version not recorded; not evidence for the current model or config.

STAGE 2PPO · Stable-Baselines3 · historical

Locomotion

Learn forward walking/running

unverified artifact; backend version not recorded; not evidence for the current model or config.

STAGE 3PPO · Stable-Baselines3 · historical

Strike

Sprint and strike prey with sickle claw

unverified artifact; backend version not recorded; not evidence for the current model or config.

Project Roadmap

PHASE 0 — v0.2.0

Clean Slate

  • TOML configs
  • Gymnasium registration
  • Developer tooling
  • Testing improvements
COMPLETE
PHASE 1 — v0.3.0

First Steps

  • Curriculum manager
  • W&B tracking
  • Locomotion metrics
  • Species training runs
IN PROGRESS
PHASE 2 — v0.4.0

Into the Wild

  • Domain randomization — planned
  • Terrain diversity
  • Turning & steering
PLANNED
PHASE 3 — v0.5.0

Evolution

  • Custom policy networks
  • New species
  • Benchmark suite
PLANNED
PHASE 4 — v0.6.0

The Pack

  • Multi-agent envs
  • Cooperative hunting
  • Predator-prey
PLANNED
PHASE 5

Hyperdrive

  • JAX/MJX backend
  • GPU benchmarks
  • Backend validation
IN PROGRESS
PHASE 6

Sim-to-Real

  • Hardware prototype — planned
  • ROS 2 bridge — planned
  • Sim-to-real validation — not started
PLANNED

Ready to Train a Dinosaur-Inspired Agent?

Explore the docs, train your first model, or contribute to the project.

GET STARTEDVIEW ON GITHUB (opens in new tab)