Task-Driven Computational Framework for Simultaneously Optimizing Design and Mounted Pose of Modular Reconfigurable Manipulators

TL;DR

The paper tackles the problem of simultaneously optimizing a modular manipulator's morphology and mounted pose for task-specific performance. It introduces a mapping function that transforms a discrete morphology state into a continuous module state $\boldsymbol{M}$, enabling unified optimization with the mounted pose $\boldsymbol{P}_m$ using Covariance Matrix Adaptation Evolutionary Strategy (CMA-ES) within a Model Predictive Control (MPC) framework. Key contributions include the continuous morphology representation, a two-level optimization that prioritizes feasibility (dynamic limits and collision avoidance) before performance (manipulability vs. joint effort), and a comprehensive comparison against genetic algorithms. Real-world drilling experiments with the CONCERT platform demonstrate practical viability and show that the proposed approach achieves faster convergence and higher-quality solutions while maintaining safety and task accuracy.

Abstract

Modular reconfigurable manipulators enable quick adaptation and versatility to address different application environments and tailor to the specific requirements of the tasks. Task performance significantly depends on the manipulator's mounted pose and morphology design, therefore posing the need of methodologies for selecting suitable modular robot configurations and mounted pose that can address the specific task requirements and required performance. Morphological changes in modular robots can be derived through a discrete optimization process involving the selective addition or removal of modules. In contrast, the adjustment of the mounted pose operates within a continuous space, allowing for smooth and precise alterations in both orientation and position. This work introduces a computational framework that simultaneously optimizes modular manipulators' mounted pose and morphology. The core of the work is that we design a mapping function that \textit{implicitly} captures the morphological state of manipulators in the continuous space. This transformation function unifies the optimization of mounted pose and morphology within a continuous space. Furthermore, our optimization framework incorporates a array of performance metrics, such as minimum joint effort and maximum manipulability, and considerations for trajectory execution error and physical and safety constraints. To highlight our method's benefits, we compare it with previous methods that framed such problem as a combinatorial optimization problem and demonstrate its practicality in selecting the modular robot configuration for executing a drilling task with the CONCERT modular robotic platform.

Figures (6)

• Figure 1: Schematic for the computational framework
• Figure 2: Top left: Encoding of modules by numbers, illustrated with a morphology state example. Middle left: An example of mapping modular states to morphology, where the red word indicates the state variable for different modules. Among them, the 9th module's state variable is the highest, followed by the 2nd module's state variable as the second highest. Bottom left: Representation of the collision model. Right: Types of modules with "in" indicating connection from the previous module and "out" showing the connection channel for subsequent modules.
• Figure 3: Arrangement of drilling points relative to the wall and world frame.
• Figure 4: Minimum cost evolution over generations with different population.
• Figure 5: The derived manipulator morphologies for the drilling task