Soft robots show compliance and have infinite degrees of freedom. Thanks to these properties, such robots are leveraged for surgery, rehabilitation, biomimetics, unstructured environment exploring, and industrial gripper. In this case, they attract scholars from a variety of areas. However, nonlinearity and hysteresis effects also bring a burden to robot modeling. Moreover, following their flexibility and adaptation, soft robot control is more challenging than rigid robot control. In order to model and control soft robots, a large number of data models are utilized in pairs or separately. This review classifies these applied data models into five kinds, which are the Jacobian model, analytical model, statistical model, neural network, and reinforcement learning, and compares the modeling and controller features, e.g., model dynamics, data requirement, and target task, within and among these categories. A discussion about the development of the existing modeling and control approaches is presented, and we forecast that the combination of offline-trained and online-learning controllers will be the widespread implementation in the future.

翻译：软体机器人具有柔顺性并拥有无限自由度。凭借这些特性，该类机器人被广泛应用于外科手术、康复医疗、仿生学、非结构化环境探索及工业夹持等领域。因此，它们吸引了众多领域学者的关注。然而，非线性与迟滞效应同时给机器人建模带来沉重负担。此外，因其灵活性与自适应能力，软体机器人的控制比刚性机器人更具挑战性。为实现软体机器人的建模与控制，大量数据模型被单独或成对使用。本综述将这些应用的数据模型分为五类：雅可比模型、解析模型、统计模型、神经网络及强化学习，并从模型动力学、数据需求、目标任务等维度对各类别内部及跨类别的建模与控制器特性进行比较。本文对现有建模与控制方法的发展进行讨论，并预测未来将广泛实现离线训练与在线学习控制器的结合应用。