Dexterous grasping is one of the most fundamental abilities of robots to implement various manipulation tasks. Robots should have the same ability as humans to plan various grasp types for dexterous grasping. This paper addresses the problem of the adaptability of grasp planning. A novel adaptive grasp planning framework is designed to adapt to various grasp types rather than a single one. In this framework, six commonly used grasp types are considered. The information of grasp type is extracted from visual data. Then, inspired by the opposition concept, a novel concept of pregrasping opposition is introduced as the pregrasping configuration to encode the information of the grasp type. After that, a two-stage adaptive grasp planning method is proposed, which determines the pregrasping opposition in Stage One and finds a feasible grasp configuration for object grasping in Stage Two. The pregrasping opposition is used as a waypoint for the formation of complex grasps. The effectiveness of the proposed framework was evaluated in simulation and real-world experiments. The experimental results demonstrated that the proposed framework can plan various grasp types for dexterous robotic grasping. Additionally, the use of grasp types helps to reduce the complexity of grasp planning and improve the grasp dexterity of robotic hands.

灵巧的抓握是机器人执行各种操纵任务的最基本能力之一。机器人应该具有与人类相同的能力，以计划各种抓握类型以进行灵巧的抓握。本文解决了抓握计划的适应性问题。一种新颖的自适应抓握计划框架旨在适应各种抓握类型，而不是单个抓握类型。在此框架中，考虑了六种常用的抓取类型。抓握类型的信息是从视觉数据中提取的。然后，在反对派概念的启发下，引入了一种预抓取反对派的新概念作为预抓取配置，以对抓取类型的信息进行编码。此后，提出了一种两阶段的自适应掌握计划方法，该方法确定了在第一阶段预抓反对派，并在第二阶段找到一种可行的抓握构形，用于抓握物体。所述pregrasping反对被用作用于掌握复杂的形成的航点。在仿真和实际实验中评估了所提出框架的有效性。实验结果表明，提出的框架可以为灵巧机器人的抓取规划各种抓取类型。另外，抓握类型的使用有助于减少抓握计划的复杂性并提高机械手的抓握灵​​巧性。