Wheeled Humanoid Bilateral Teleoperation with Position-Force Control Modes for Dynamic Loco-Manipulation

TL;DR

The paper tackles Dynamic Loco-Manipulation by developing a bilateral teleoperation framework with loco-manipulation retargeting and hybrid position-force control modes. It introduces Locomotion Retargeting, Manipulation Retargeting, and whole-body Moment Feedback to enable intuitive, immersive control of a wheeled humanoid for heavy-object manipulation and collaborative tasks. Key contributions include: a dual-mode (precision and dynamic) mapping for both locomotion and manipulation, an IK-based manipulation retargeting with impedance options, and experimental validation on box slotting and leader/follower HRC, with insights on moment feedback usefulness. The results demonstrate that hybrid control modes support efficient, compliant DLM in challenging tasks, with potential impact on industrial automation, construction, and healthcare robotics.

Abstract

Remote-controlled humanoid robots can revolutionize manufacturing, construction, and healthcare industries by performing complex or dangerous manual tasks traditionally done by humans. We refer to these behaviors as Dynamic Loco-Manipulation (DLM). To successfully complete these tasks, humans control the position of their bodies and contact forces at their hands. To enable similar whole-body control in humanoids, we introduce loco-manipulation retargeting strategies with switched position and force control modes in a bilateral teleoperation framework. Our proposed locomotion mappings use the pitch and yaw of the operator's torso to control robot position or acceleration. The manipulation retargeting maps the operator's arm movements to the robot's arms for joint-position or impedance control of the end-effector. A Human-Machine Interface captures the teleoperator's motion and provides haptic feedback to their torso, enhancing their awareness of the robot's interactions with the environment. In this paper, we demonstrate two forms of DLM. First, we show the robot slotting heavy boxes (5-10.5 kg), weighing up to 83% of the robot's weight, into desired positions. Second, we show human-robot collaboration for carrying an object, where the robot and teleoperator take on leader and follower roles.

Figures (7)

• Figure 1: The teleoperator (left) controls whole-body of the robot to accomplish a collaborative carrying task with a human. The human machine interface provides haptic feedback and applies a wrench on the human body, synchronizing the robot and teleoperator body motion.
• Figure 2: The reduced order models for the human and robot with their corresponding states are shown. The human and robot heights are given by $h_H$ and $h_R$, respectively. Using the forceplate, we are able to measure the ground reaction forces (GRF) and estimate the human's applied moment on the ground, $M_{zH}$.
• Figure 3: Full bilateral teleoperation layout and control framework. The pilot chooses 1 of 2 mappings for each of the control modes: sagittal locomotion, yaw locomotion, and arm manipulation. The haptic feedback to the pilot consists of force and moment feedback to synchronize human-robot motion and improve the teleoperator's estimate of the robot's states and of contact with the environment.
• Figure 4: The table summarizes the teleoperator's control modes, highlighting the option to choose between precise position or dynamic force/acceleration mappings for robot locomotion and manipulation.
• Figure 5: (A) displays the robot odometry for two box-pushing trials with different end locations: Destination 1 and 2. (B) shows the sagittal locomotion switching behavior where the pilot uses precise velocity mode (in blue) to align the robot pre-contact, and the dynamic acceleration mode (in orange) while pushing. (C) shows the yaw locomotion modes used and highlights the pilots preference for precise position control (in blue) during the majority of the task. The graphic in (D) and the green area showing the arm $(x,y)$ position in (E), highlight the pilot strategy for rotating the heavy-box by moving the left and right hands. (F) shows the contact forces at each hand and the rapid changes in contact as sensed by the contact sensor at each hand.