Description
Hello,
I've been working with the [demo] project and encountered a couple of issues that I need some guidance on. I successfully converted initial_configs.p and target_object_names.p in the benchmark folder from .p to .json, modified them, and converted them back to .p, which worked fine. However, I'm facing issues with the .p files in other folders, such as robot0.
Issue with Converting .p Files in robot0 Folder:
I tried a similar approach to convert and modify .p files in the robot0 folder but encountered errors during execution. I suspect that something might be getting lost or misinterpreted during the conversion process. Could you provide insights on what the .p files in these folders should contain or if there's a specific structure or content that I need to maintain while modifying these files?
Customizing Task Runners for Specific Tasks:
Additionally, I am interested in using this model to plan tasks where only the initial state and goal pose are provided, without the need for grasping and manipulating objects. Is it possible to modify the prepare_task_runners function to configure tasks based solely on initial states and goal poses? If yes, could you provide some guidance or examples on how to achieve this?
Any assistance or pointers on these matters would be greatly appreciated. Thank you for your support and for the great work on this project.