New method generates reliable repair code for assembly faults
FORGE: Fault-Oriented Rework Code Generation for Executable Interlocking Assembly
Artificial Intelligence
Summary
Sometimes machines or structures get broken during assembly, and fixing them without causing more problems is tricky. This paper shows that just writing a quick repair code based on the problem often fails because it misses key steps or dependencies. The authors created a new approach called FORGE that carefully plans the repair steps and turns them into actual executable code that restores the assembly correctly. They also made a big set of test problems to measure how well such repair code can be generated and tested.
What this means in practice
- •For industrial automation engineers: Produce executable repair sequences that can fix faults in complex assembly processes without causing new errors.
- •For robotics software developers: Develop control programs that automatically generate fault recovery routines for robotic assembly tasks using structured reasoning.
Authors
Fuqiang Wang, Jiahao Fu, Ren Hefeng, Yining Qian, Chuanting Zhang, Cheng Tan, Yuwen Liu, Jingxuan Wei
Abstract
Large language models have demonstrated strong capabilities in structured reasoning and code generation, yet identifying an assembly fault does not guarantee the ability to generate executable rework programs. In interlocking assemblies, a locally plausible repair program may violate blocking dependencies, omit required recovery operations, or terminate in an incorrect configuration. We identify this challenge as fault-conditioned executable rework generation, where models must jointly reason about structural failures and generate programs that restore constrained assembly states. To benchmark this capability, we introduce FORGE-Bench, containing 4,000 tasks across 26 interlocking assemblies with five assemblies held out for evaluating generalization to unseen structures. We further propose FORGE (Fault-Oriented Rework Code Generation for Executable Interlocking Assembly), a structured state-to-program framework that decomposes rework generation into Rework Structuring and Action Grounding. Rework Structuring captures fault-induced recovery dependencies, while Action Grounding maps these dependencies into state-conditioned executable actions. Experiments show that locally reasonable repair code is insufficient for reliable assembly recovery, and that connecting structural recovery reasoning with executable action realization is essential for verified target restoration.