Adaptive Pushing with a Balancing Mobile Manipulator

Abstract

Dynamic mobile manipulation—the ability of mobile robots to forcefully interact with their environments—remains a core challenge for deploying robots in human-centered spaces. This thesis investigates this challenge through a focused case study on dynamic pushing, where a dynamically balancing robot manipulates heavy or constrained objects via pushing. The goal is to derive insights that can inform the development of more adaptive and robust mobile manipulation algorithms and systems.

First, we investigate effective wheelchair maneuvering using a dynamically balancing mobile manipulator. Wheelchairs represent a type of nonholonomic cart system, maneuvering such systems with mobile manipulators~(MM) is challenging mostly due to the following reasons: 1. These systems feature nonholonomic constraints and considerably varying inertial parameters that require online identification and adaptation. 2. These systems are widely used in human-centered environments, which demand the MM to operate in potentially crowded spaces while ensuring compliance for safe physical human-robot interaction~(pHRI). We propose a control framework that plans whole-body motion based on quasi-static analysis to maneuver heavy nonholonomic carts while maintaining overall compliance. We validated our approach experimentally by maneuvering a wheelchair with a bimanual mobile manipulator, the CMU ballbot. The experiments demonstrate the proposed framework is able to track desired wheelchair velocity with loads varying from 11.8kg to 79.4kg at a maximum linear velocity of 0.45~m/s and angular velocity of 0.3~rad/s. Furthermore, we verified that the proposed method can generate human-like motion smoothness of the wheelchair while ensuring safe interactions with the environment.

Second, we explore a more generalizable dynamic non-prehensile pushing algorithm for manipulating unknown objects. In this setting, a key challenge is to manipulate objects with unknown dynamics which are difficult to infer from visual observation. To address this, we propose an adaptive dynamics model for common movable indoor objects via a learned $SE(2)$ dynamics representation. This model is integrated into Model Predictive Path Integral (MPPI) control to guide the robot's motion. Additionally, the learned dynamics help inform decision-making when navigating around objects that cannot be manipulated. Our approach is validated in both simulation and real-world scenarios, demonstrating its ability to accurately represent object dynamics and effectively manipulate various objects. We further highlight its success in the Navigation Among Movable Objects (NAMO) task by deploying the proposed framework on a dynamically balancing mobile robot, Shmoobot.

Finally, we discuss the development of the CMU Ballbot and CMU Shmoobot, both ball-balancing mobile manipulators that serve as experimental testbeds throughout this thesis—and the insights gained from their use in more general dynamic mobile manipulation tasks.