Collaborative Manipulation of Deformable Objects with Predictive Obstacle Avoidance

TL;DR

The paper tackles real-time collaborative manipulation of deformable objects by multiple robots, formulating a lead-follow task with safety constraints. It integrates Position Based Dynamics for fast state prediction with Control Barrier Functions encoded as a QP-based safe controller, enabling collision and overstretching avoidance in a multi-agent setting and implemented in ROS. Key contributions include a unified simulation-and-control framework, object-state–aware safety constraints with real-time Jacobian estimation, and demonstration across 1D and 2D deformable objects with up to three assistants, achieving sub-10 ms QP solve times. This approach advances practical DOM by providing real-time safety guarantees and scalable multi-agent coordination, with potential impact on industrial and domestic manipulation tasks. The work lays groundwork for physical-system validation and extensions to dynamic environments and holding-point orientations.

Abstract

Manipulating deformable objects arises in daily life and numerous applications. Despite phenomenal advances in industrial robotics, manipulation of deformable objects remains mostly a manual task. This is because of the high number of internal degrees of freedom and the complexity of predicting its motion. In this paper, we apply the computationally efficient position-based dynamics method to predict object motion and distance to obstacles. This distance is incorporated in a control barrier function for the resolved motion kinematic control for one or more robots to adjust their motion to avoid colliding with the obstacles. The controller has been applied in simulations to 1D and 2D deformable objects with varying numbers of assistant agents, demonstrating its versatility across different object types and multi-agent systems. Results indicate the feasibility of real-time collision avoidance through deformable object simulation, minimizing path tracking error while maintaining a predefined minimum distance from obstacles and preventing overstretching of the deformable object. The implementation is performed in ROS, allowing ready portability to different applications.

Figures (9)

• Figure 1: Example deformable material handling tasks.
• Figure 2: Software architecture of the proposed method.
• Figure 3: Overlaid snapshots of rope-like object with a single assistant simulation at ($t=30$, $60$, $130$, $160$ s). The assistant aims to maintain its initial relative position to the leader, indicated by the arrow, while ensuring object safety and preventing collisions with the grey ground.
• Figure 4: Overlaid snapshots of stiff rod object with single assistant simulation at ($t=30$, $60$, $120$, $150$ s).
• Figure 5: Data plot of the controllers in the single assistant agent simulations.