Reinforcement Learning Research Kit

Push the boundaries of Spot’s mobility with our new package for exploring AI-based capabilities

The Reinforcement Learning (RL) Research Kit provides joint-level API access to broaden research possibilities—from faster gaits to more complex movements. Train in a ready-to-use simulation environment, then implement in the real world with the assurance of premium hardware servicing.

Features

Create advanced skills

Get lower level access to Spot to explore new behaviors and capabilities

Get started quickly

Access everything you need out-of-the-box to keep innovating with Spot

Explore more freely

Protect your robot with a custom simulator and premium service

Specifications

SPOT RESEARCHER API

Joint-level API

Built and maintained by Boston Dynamics

Documentation

Lightweight developer documentation (developer support not included)

SPOT SIMULATOR

Open source model

Dynamic model of Spot (including masses, inertias, and actuator models) and a mimic of the Spot Researcher API available for download in NVIDIA Isaac Lab

DEVELOPER HARDWARE

Payload mount

Brackets and cables to attach the NVIDIA AGX Orin Developer Kit—recommended for optimal computing speed and latency—to Spot (AGX Orin sold separately)

SERVICE & SUPPORT

Spot Care

Premium package to ensure your robot continues operating at its best and can get back to work as quickly and smoothly as possible after an accident

Get started with the RL Research Kit

With Spot, we provide a flexible platform for mobility and manipulation. Now it’s your turn to change how we think about robotics.