Reconfigurable Robot Control Using Flexible Coupling Mechanisms

TL;DR

This work tackles the problem of energy-efficient, robust coupling in self-reconfigurable robot swarms by introducing a soft asymmetric anchor that enables easy coupling yet strong holding forces. It couples this passive mechanism with a Model Predictive Control framework that enforces polygon-based constraints to maintain the geometry of a connected assembly while allowing flexible motions. The main contributions are the soft anchor design with a quantified force profile, a three-bar linkage simulation model, and a MCP-based, polygon-constrained control scheme validated through both Bullet simulations and real hardware on a ROS-based platform. The approach supports forming flexible chains and bridges, decoupling on demand, and navigating unstructured terrains, offering practical implications for terrain adaptation and collaborative robotic infrastructure. Looking ahead, the authors propose decentralized MPC for scalability and exploration of anchor scalability and onboard sensing for environment-aware assembly.

Abstract

Reconfigurable robot swarms are capable of connecting with each other to form complex structures. Current mechanical or magnetic connection mechanisms can be complicated to manufacture, consume high power, have a limited load-bearing capacity, or can only form rigid structures. In this paper, we present our low-cost soft anchor design that enables flexible coupling and decoupling between robots. Our asymmetric anchor requires minimal force to be pushed into the opening of another robot while having a strong pulling force so that the connection between robots can be secured. To maintain this flexible coupling mechanism as an assembled structure, we present our Model Predictive Control (MPC) frameworks with polygon constraints to model the geometric relationship between robots. We conducted experiments on the soft anchor to obtain its force profile, which informed the three-bar linkage model of the anchor in the simulations. We show that the proposed mechanism and MPC frameworks enable the robots to couple, decouple, and perform various behaviors in both the simulation environment and hardware platform. Our code is available at https://github.com/ZoomLabCMU/puzzlebot_anchor . Video is available at https://www.youtube.com/watch?v=R3gFplorCJg .

Figures (10)

• Figure 1: Top: Robots can form a flexible chain and go down a step (left) or a rigid structure that crosses a gap between two equal-height test stages (right). Bottom left: Four robots forming an articulated chain. Bottom right: One robot inserts its anchor inside the opening of another one (circuitry removed for visual purposes).
• Figure 2: (a) Anchor design. (b) Anchor (gray) with its base inserted in a holder (blue) acting as a floating joint. (c) Anchor and an opening uncoupled when on the same height. (d) Anchor and opening coupled. The anchor tips are sitting in the slits. (e) Anchor locks the connection when the robot's vertical position changes with gravity due to the irregularities on the ground.
• Figure 3: States of the 3D printed anchor from two viewpoints. (a) Anchor collapsed. In this state, the anchor is in its shortest configuration. (b) Anchor extended. The holder provides translational motion within 5 mm. (c) Anchor rotated. When not collapsed, the anchor is also free to perform rotations within 0.5 rad.
• Figure 4: (a) Two robots coupled together with an extended anchor. The dark gray area in the middle of the robot is the battery box that extends to both the upper slits and the lower opening. This limits the anchor state and enforces the anchor tip to sit and lock with the opening. (b) One robot rotates as the anchor is coupled. The white shaded anchor shows a projection of the real anchor if the anchor joint is not compliant. (c) The blue shaded area shows the constraint of anchor head location during coupling.
• Figure 5: A point inside a convex polygon.