Can Moveit-Python Interface be used to simulate actions taken by RL agent in Gazebo?
I am doing a Project to coordinate two Universal Robots (UR5) using Reinforcement Learning Algorithms (Preferably by PPO i.e. Proximal Policy Optimization Technique). My question can Moveit - Python Interface be used for simulating the actions taken by the RL Agent in Gazebo? Moveit has an advantage that it has predefined FK (forward kinematics) and IK (inverse kinematics) solvers which would save a lot of coding. But is it recommended to use Moveit? Because I will be dealing with continuous state space and action spaces.