ManiDreams: An Open-Source Library for Robust Object Manipulation via Uncertainty-aware Task-specific Intuitive Physics

TL;DR

ManiDreams is an open-source framework that enhances robotic manipulation robustness by integrating uncertainty representation and propagation into physics-based planning, improving performance under real-world perturbations.

Contribution

It introduces a modular, uncertainty-aware manipulation planning framework that explicitly models and propagates uncertainties, enabling robust manipulation without retraining existing policies.

Findings

Maintains robust performance under various perturbations.

Flexible across different policies, optimizers, and physics backends.

Demonstrates effectiveness in real-world manipulation tasks.

Abstract

Dynamics models, whether simulators or learned world models, have long been central to robotic manipulation, but most focus on minimizing prediction error rather than confronting a more fundamental challenge: real-world manipulation is inherently uncertain. We argue that robust manipulation under uncertainty is fundamentally an integration problem: uncertainties must be represented, propagated, and constrained within the planning loop, not merely suppressed during training. We present and open-source ManiDreams, a modular framework for uncertainty-aware manipulation planning over intuitive physics models. It realizes this integration through composable abstractions for distributional state representation, backend-agnostic dynamics prediction, and declarative constraint specification for action optimization. The framework explicitly addresses three sources of uncertainty: perceptual, parametric, and structural. It wraps any base policy with a sample-predict-constrain loop that evaluates candidate actions against distributional outcomes, adding robustness without retraining. Experiments on ManiSkill tasks show that ManiDreams maintains robust performance under various perturbations where the RL baseline degrades significantly. Runnable examples on pushing, picking, catching, and real-world deployment demonstrate flexibility across different policies, optimizers, physics backends, and executors. The framework is publicly available at https://github.com/Rice-RobotPI-Lab/ManiDreams