Custom Environments and Algorithms

rlbotics is not tied to a simulator or to PPO - implement the VecEnv interface and the runner trains against it, or pass your own algorithm class to the runner.

TIP

Install with pip install telekinesis-rlbotics — see Install Support for Reinforcement Learning for the simulator backends.

Your Own Environment

VecEnv is three methods and a handful of attributes. The three built-in adapters for Gymnasium, mjlab and Isaac Lab are each about 200 lines and are the reference implementations.

RLBotics

Introduce a Custom Environment

Implements the VecEnv interface - observation groups, time-limit bootstrapping, action bounds, rendering and validation.

View →

Your Own Algorithm

The runner owns the loop and calls into the algorithm through a fixed set of methods, so anything implementing them can take PPO's place.

RLBotics

Introduce a Custom Algorithm

Subclasses PPO or replaces it, adds hyperparameters through a config, and wires the class into the runner.

View →

Where to Go Next?

Ready to go deeper? Browse the Telekinesis Skill Examples repository for a runnable example covering every Skill.

Skill Examples

One runnable example per Skill, across vision, 3D, hardware, and robotics.

Explore →

Support

Join Discord

Ask questions, share what you're building, and get help from the community.

Join →

Report a bug?

Found a bug or something not working as expected? Raise an issue on our GitHub.

Raise an issue →