Learning to Play Cup-and-Ball with Noisy Camera Observations

TL;DR

This paper introduces a learning-based control strategy for a robot to catch a ball in a cup using noisy camera data, combining offline planning and online convex optimization, with an iterative noise support learning framework.

Contribution

It presents a novel iterative framework for learning camera noise support to improve robotic ball-catching performance under noisy observations.

Findings

Probability of successful catch increases as noise support learning converges.

Simulation and experimental results validate the theoretical guarantees.

The approach effectively handles noisy sensor data in real-time control.

Abstract

Playing the cup-and-ball game is an intriguing task for robotics research since it abstracts important problem characteristics including system nonlinearity, contact forces and precise positioning as terminal goal. In this paper, we present a learning model based control strategy for the cup-and-ball game, where a Universal Robots UR5e manipulator arm learns to catch a ball in one of the cups on a Kendama. Our control problem is divided into two sub-tasks, namely $(i)$ swinging the ball up in a constrained motion, and $(ii)$ catching the free-falling ball. The swing-up trajectory is computed offline, and applied in open-loop to the arm. Subsequently, a convex optimization problem is solved online during the ball's free-fall to control the manipulator and catch the ball. The controller utilizes noisy position feedback of the ball from an Intel RealSense D435 depth camera. We propose a novel iterative framework, where data is used to learn the support of the camera noise distribution iteratively in order to update the control policy. The probability of a catch with a fixed policy is computed empirically with a user specified number of roll-outs. Our design guarantees that probability of the catch increases in the limit, as the learned support nears the true support of the camera noise distribution. High-fidelity Mujoco simulations and preliminary experimental results support our theoretical analysis.