2. Hardware and simulation environment
3. Define tasks and reward functions
4. Reinforcement learning algorithm selection
5. Training the robot
6. Assessment and Deployment