Creating machines capable of directly interacting with the environment is a critical challenge in robotics. Research on robot manipulation aiming to produce machines capable of directly and autonomously interacting with the environment has increased in the last decades. Learning will be pivotal to such autonomous systems. The EU-funded IntelliMan project will focus on how a robot can efficiently learn to manipulate in a purposeful and highly performant way. The project will develop an innovative AI-Powered Manipulation System with persistent learning capabilities, able to perceive characteristics and features of the environment through a heterogeneous set of sensors, decide how to execute a task autonomously, detect failures in the task execution, and request new knowledge.