Is there any documentation on integrating a real time example to help train a robot with reinforcement learning?

[Isaacsimfyp1]

We are currently working on our Final Year Project that is based on training an upper humanoid to do some tasks (focusing on picking up a cup with one arm right now). Initially, we were training the model on pure trial and error basis through RL but the training takes too long and the results are nowhere near fruitful given how much time we have left (barely few months before we graduate). Recently, an idea sparked to sort of use "imitation learning" concepts and use a real time example of a human picking a cup from the table (we do have a motion capture set up) and using that example as our reference/starting data and building our policy on it. However, I cannot find any documentation on it or anything on the forums that could help us in detail. The main issue is integration of our real time data (we have it in an excel sheet format). Can anyone guide me in this regard?