Task Space Control of Hydraulic Construction Machines Using Reinforcement Learning

Teleoperation is vital in the construction industry, allowing safe machine manipulation from a distance. However, controlling machines at a joint level requires extensive training due to their complex degrees of freedom. Task space control offers intuitive maneuvering, but precise control often requires dynamic models, posing challenges for hydraulic machines. To address this, we use a data-driven actuator model to capture machine dynamics in real-world operations. By integrating this model into simulation and reinforcement learning, a control policy for task space control is obtained. A 3t hydraulic construction machine, Brokk 170, serves as the platform for implementing the proposed approach. Through a series of experiments, the framework's validity is established by comparing it against a well-established Jacobian-based approach.