COBRA-PPM: A Causal Bayesian Reasoning Architecture Using Probabilistic Programming for Robot Manipulation Under Uncertainty

TL;DR

COBRA-PPM introduces a causal Bayesian reasoning framework using probabilistic programming, enabling robots to better understand and predict manipulation outcomes under uncertainty, demonstrated through high success rates in simulated and real-world tasks.

Contribution

It combines causal Bayesian networks with probabilistic programming to perform interventional inference, advancing robot manipulation reasoning under uncertainty.

Findings

Achieved 88.6% prediction accuracy in simulation.

Attained 94.2% task success rate in manipulation tasks.

Effective sim2real transfer on a domestic robot.

Abstract

Manipulation tasks require robots to reason about cause and effect when interacting with objects. Yet, many data-driven approaches lack causal semantics and thus only consider correlations. We introduce COBRA-PPM, a novel causal Bayesian reasoning architecture that combines causal Bayesian networks and probabilistic programming to perform interventional inference for robot manipulation under uncertainty. We demonstrate its capabilities through high-fidelity Gazebo-based experiments on an exemplar block stacking task, where it predicts manipulation outcomes with high accuracy (Pred Acc: 88.6%) and performs greedy next-best action selection with a 94.2% task success rate. We further demonstrate sim2real transfer on a domestic robot, showing effectiveness in handling real-world uncertainty from sensor noise and stochastic actions. Our generalised and extensible framework supports a wide range of manipulation scenarios and lays a foundation for future work at the intersection of robotics and causality.