Catch, Throw, Repeat: Planning for Human-Robot Partner Juggling

Jonathan Rainer Lippert, Kai Ploeger, Abir Chowdhury, Hermann Müller, Jan Peters, Alap Kshirsagar · N/A · 2026

Summary

Dynamic object exchange between humans and robots remains a challenging problem due to uncertainty in perception, timing, and contact-rich interaction.

Abstract Summary

Dynamic object exchange between humans and robots remains a challenging problem due to uncertainty in perception, timing, and contact-rich interaction. Human-robot juggling represents a particularly demanding instance of this problem, requiring precise real-time coordination, predictive motion planning with feedback control, and robustness to variability in human motion. Enabling such skills is of interest for advancing physical human-robot interaction and shared autonomy. We present a real-time planning and control architecture for human-robot partner juggling that enables a robot to reliably catch and throw balls in synchronized multi-ball patterns with a human partner. The system integrates predictive ball tracking, adaptive online trajectory optimization using a multiple-shooting formulation, and a state-machine-based coordination logic to enable synchronized multi-ball human-robot partner juggling. In a user study with 8 participants of varying juggling skill from beginner to expert, we demonstrate that our system can achieve three-ball cascades shared between the robot and the human. All participants exceeded previously reported best-case results within a 10-minute test session, with one participant extending the previous record for shared three-ball cascade juggling fivefold to 20 consecutive robot catches, and another participant achieving a 100% success rate with 40 consecutive catches in a single-ball catch-and-return setting. Video documentation can be found at https://kai-ploeger.com/partner-juggling

Key Points

Dynamic object exchange between humans and robots remains a challenging problem due to uncertainty in perception, timing, and contact-rich interaction.
Human-robot juggling represents a particularly demanding instance of this problem, requiring precise real-time coordination, predictive motion planning with feedback control, and robustness to variability in human motion.
Enabling such skills is of interest for advancing physical human-robot interaction and shared autonomy.
We present a real-time planning and control architecture for human-robot partner juggling that enables a robot to reliably catch and throw balls in synchronized multi-ball patterns with a human partner.
The system integrates predictive ball tracking, adaptive online trajectory optimization using a multiple-shooting formulation, and a state-machine-based coordination logic to enable synchronized multi-ball human-robot partner juggling.

Method

We approach the task of human-robot partner juggling by cascading a ball tracking and prediction pipeline, and a flexible trajectory planner capable of real-time online replanning, using a state machine based coordination logic . Ball positions are observed using an OptiTrack marker-based motion capture system, treating each ball as an identical point marker.

At each time step, position measurements are associated with individual ball tracks using a Hungarian assignment over Mahalanobis distances.

Method — Catch, Throw, Repeat: Planning for Human-Robot Partner Juggling — Fig. 2: Post-release constraint geometry. The end-effector motion is restricted such that the predicted ball trajectory remains within the green cone-shaped region originating at the end-effector frame origin, preventing unintended contact with the ball after takeoff. During the carry phase, the ball rests at the frame origin.

Experiments

To evaluate the proposed approach, we conduct experiments using a $160\text{\,}\mathrm{V}$ dual-encoder Barrett WAM robotic manipulator, configured with four degrees of freedom corresponding to the shoulder and elbow joints and mounted in an orientation that results in anthropomorphic arm kinematics, as shown in Fig. 1 .

The cable-driven, backdrivable actuation enables accurate torque control and is well suited for dynamic catching and throwing motions.

Experiments — Catch, Throw, Repeat: Planning for Human-Robot Partner Juggling — Fig. 3: RViz screenshot of the simulated robot catching a ball within its reachable workspace in the horizontal catching plane. Touchdown locations are predicted from a least-squares fit to recent ball measurements.

Sources and Demonstrations

Figures are reproduced from Lippert et al. See the full paper for experimental details and the project page, when the paper links one, for demonstrations and videos.