Evaluation of Frameworks That Combine Evolution and Learning to Design Robots in Complex Morphological Spaces
Evaluation of Frameworks That Combine Evolution and Learning to Design Robots in Complex Morphological Spaces
复制标题
复杂形态空间中结合进化和学习设计机器人的框架评估
DOI：
10.1109/tevc.2023.3316363
复制
发表时间：
2023
影响因子：
14.3
通讯作者：
Li W
中科院分区：
文献类型：
--
作者：
Li W
众所周知，联合优化机器人的身体和大脑是一项具有挑战性的任务，特别是当试图在随后将在真实的世界中构建的模拟中进化设计时。为了解决这个问题，越来越普遍的做法是将联合收割机进化与学习算法相结合，这种算法可以改进新后代的遗传控制器，使其适应新的车身设计，也可以从头开始学习。在这篇文章中，提出了一种方法，其中一个机器人是间接指定的两个组成模式产生网络（CPPN）编码在一个单一的基因组，一个编码的大脑和其他的身体。使用进化算法（EA）进化基因组的身体部分，与个人学习算法（也是EA）应用于继承的控制器，以改善它。本文的目标是确定如何最有效地利用学习过程的结果，以提高机器人的任务性能。具体而言，研究了三种变体：1）仅身体+控制器的进化; 2）将学习算法应用于遗传控制器，并将学习到的适应度分配给基因组;以及3）应用学习，并使用学习到的控制器更新基因组，以及分配学习到的适应度。实验在三种不同的情况下进行，选择有利于不同的机构和运动模式。结果表明，更好的性能，可以使用学习，但只有当学习控制器是由后代继承。
Jointly optimizing both the body and brain of a robot is known to be a challenging task, especially when attempting to evolve designs in simulation that will subsequently be built in the real world. To address this, it is increasingly common to combine evolution with a learning algorithm that can either improve the inherited controllers of new offspring to fine tune them to the new body design or learn them from scratch. In this article an approach is proposed in which a robot is specified indirectly by two compositional pattern producing networks (CPPNs) encoded in a single genome, one which encodes the brain and the other the body. The body part of the genome is evolved using an evolutionary algorithm (EA), with an individual learning algorithm (also an EA) applied to the inherited controller to improve it. The goal of this article is to determine how to utilize the results of learning process most effectively to improve task performance of the robot. Specifically, three variants are investigated: 1) evolution of the body+controller only; 2) a learning algorithm is applied to the inherited controller with the learned fitness assigned to the genome; and 3) learning is applied and the genome is updated with the learned controller, as well as being assigned the learned fitness. Experiments are performed in three different scenarios chosen to favor different bodies and locomotion patterns. It is shown that better performance can be obtained using learning but only if the learned controller is inherited by the offspring.