Table 2 Pros and cons of SL, LfD, and RL

SL

1. This approach is very accurate for seen clothes recognition using the

training dataset [50]

2. Given data and labels is very helpful for classification and detecting

clothes grasping points for the cloth application [64]

1. Without prior knowledge about specific garments, clothes classification and recognition is inaccurate [64]

2. It is not easy to extend to complex cloth manipulation scenarios [65]

LfD

1. This approach can transfer the learned motion to unseen scenarios [4]

2. The demonstration provides a high-level plan that is used to execute low-level control for cloth manipulation [66]

1. This approach mostly focuses on a single task, which means it is difficult to learn to manipulate new objects [66]

2. After the demonstration, this approach is not easy to adapt different scales or shape of robotic clothing assistants.[67] [29]

RL

1. This approach makes it possible to accomplish complex clothing assistance tasks using appropriate reward functions [60]

2. It is the only way to collect data for interacting with the environment [68]

1. It is still challenging to find suitable reward functions for complex cloth manipulation without reward shaping [60]

2. This approach requires a lot of data which makes the computation time consuming which, in turn, increases the difficulty to achieve the results [68]