Abstract:The work is inspired by the assembly of Soma block puzzles. Soma block puzzles usually include unstable intermediate states that require additional support to maintain stability temporarily. In the puzzles’ solution manual, we can observe that designers consider the characteristics that humans have two hands and can avoid an unstable intermediate state by using one hand to support the finished component and using the other hand to assemble an upcoming workpiece. Motivated by human behavior, this paper develops a planner that automatically finds an optimal assembly sequence for a dual-arm robot to build a woodblock structure while considering various constraints and supporting grasps from a second hand. It uses the mesh model of wood blocks and the final assembly state to generate possible assembly sequences and evaluate the optimal assembly sequence by considering the stability, graspability, assemblability, and the need for a second hand. Especially, the need for a second hand is resolved when supports from worktables and other workpieces are not enough to produce a stable assembly. A second hand can hold and support the unstable components so that the robot can further assemble new workpieces until the structure state becomes stable again. The output of the planner includes the optimal assembly orders, candidate grasps, assembly directions, and the supporting grasps (if needed). The output can help guide a dual-arm robot to perform motion planning and thus generate assembly motion. Experiments using various blocks and structures show the effectiveness of the proposed planner. Note to Practitioners—The presented planner can generate an optimal assembly order for a large variety of structures like decoration accessories, furniture, home interiors, frames, etc., in the practices. They can also be used for scenarios that need stacking or piling up multiple objects. The generated optimal assembly order is more friendly to dual-arm robot systems than previous assembly planners that ignored the merits of robotic collaboration. Also, the proposed assembly planner generates the necessary information for the motion planner, such as grasp poses and optimal assembly directions. A motion planner can directly use the generated results to plan robotic assembly motion.. The proposed assembly planner is expected to significantly reduce human effort and increase the efficiency of robotic assembly lines.

Approach To Cooperative Assembly Task Planning For Multiple Manipulators

A Collaborative Approach To Assembly Sequence Planning

Distributed Cooperative Motion Planning for Multiple Articulated Robots

Knowledge-based Intelligent Assembly Planning System and Its Related Key Technologies

Human-in-the-loop Robotic Manipulation Planning for Collaborative Assembly

Long-Horizon Multi-Robot Rearrangement Planning for Construction Assembly

Integrated Task Sequence Planning and Assignment for Human–robot Collaborative Assembly Station

A knowledge-based approach to assembly sequence planning

Preparatory Manipulation Planning using Automatically Determined Single and Dual Arms

Interactive task planning in virtual assembly.

Multi-Robot Geometric Task-and-Motion Planning for Collaborative Manipulation Tasks

A Hierarchical Approach to Disassembly Sequence Planning for Mechanical Product

A Hierarchical Human-Robot Interaction-Planning Framework for Task Allocation in Collaborative Industrial Assembly Processes

Planning to Build Block Structures With Unstable Intermediate States Using Two Manipulators

Robust Task Planning for Assembly Lines with Human-Robot Collaboration

Assembly-Oriented Task Sequence Planning for a Dual-Arm Robot

Assembly Sequence Planning for Motion Planning

Hierarchical Motion Planning Framework for Cooperative Transportation of Multiple Mobile Manipulators

A Geometric Method Based on Space Arc for Pose-Configuration Simultaneous Planning of Segmented Hyper-Redundant Manipulators

Analysis of autonomous cooperative assembly using coordination schemes by heterogeneous robots using a control basis approach

Applying Improved Particle Swarm Optimization to Asynchronous Parallel Disassembly Planning