Multi-objective symbolic regression for physics-aware dynamic modeling

作者：

Highlights：

• Using prior knowledge helps to compensate for the training data insufficiencies.

• Using prior knowledge helps to construct physically justified models.

• Model-based motion controllers do better with models learned using prior knowledge.

摘要

•Using prior knowledge helps to compensate for the training data insufficiencies.•Using prior knowledge helps to construct physically justified models.•Model-based motion controllers do better with models learned using prior knowledge.

论文关键词：Symbolic regression,Genetic programming,Model learning for control,Motion control

论文评审过程：Received 22 October 2019, Revised 12 March 2021, Accepted 13 May 2021, Available online 31 May 2021, Version of Record 3 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115210