Overview

Robotic assembly spans recognizing parts and their relationships, planning feasible sequences and motions, and controlling robots as components make contact. Common approaches combine CAD-based or symbolic planning with imitation learning and reinforcement learning; vision-language models help interpret instructions, while force, tactile, and compliant control guide precise insertion and recovery. Systems are tested on tasks such as peg-in-hole and connector insertion, NIST assembly boards, and longer furniture, industrial, and construction assemblies, often with human or multiple robots. Precise, short tasks can be reliable, but generalizing to unfamiliar parts and completing long sequences remains difficult because small perception or execution errors compound, and simulation performance may not transfer to hardware.

Latest papers

CardsList

93 papers7 papers in the last four weeks, down 59% on the four weeks before. 0.1% of all new papers.

  1. From Perception to Symbolic Task Planning: Vision-Language Guided Human-Robot Collaborative Structured Assembly

    Jan 2, 2026Yanyi Chen, Min DengHuman-Robot CollaborationRobotic Assembly

  2. Improving dependability in robotized bolting operations

    Nov 13, 2025Lorenzo Pagliara, Vincenzo Petrone, Violeta Redondo +3Industrial RoboticsRobotic Assembly

  3. IDfRA: Self-Verification for Iterative Design in Robotic Assembly

    Sep 21, 2025Nishka Khendry, Christos Margadji, Sebastian W. PattinsonRobotic AssemblyRobot Task Planning