Data-driven control of hydraulic impact hammers under strict operational and control constraints

Abstract

This paper presents a data-driven methodology for the control of static hydraulic impact hammers, also known as rock breakers, which are commonly used in the mining industry. The task addressed in this work is that of controlling the rock-breaker so its end-effector reaches arbitrary target poses, which is required in normal operation to place the hammer on top of rocks that need to be fractured. The proposed approach considers several constraints, such as unobserved state variables due to limited sensing and the strict requirement of using a discrete control interface at the joint level. First, the proposed methodology addresses the problem of system identification to obtain an approximate dynamic model of the hydraulic arm. This is done via supervised learning, using only teleoperation data. The learned dynamic model is then exploited to obtain a controller capable of reaching target end-effector poses. For policy synthesis, both reinforcement learning (RL) and model predictive control (MPC) algorithms are utilized and contrasted. As a case study, we consider the automation of a Bobcat E10 mini-excavator arm with a hydraulic impact hammer attached as end-effector. Using this machine, both the system identification and policy synthesis stages are studied in simulation and in the real world. The best RL-based policy consistently reaches target end-effector poses with position errors below 12 cm and pitch angle errors below 0.08 rad in the real world. Considering that the impact hammer has a 4 cm diameter chisel, this level of precision is sufficient for breaking rocks. Notably, this is accomplished by relying only on approximately 68 min of teleoperation data to train and 8 min to evaluate the dynamic model, and without performing any adjustments for a successful policy Sim2Real transfer. A demonstration of policy execution in the real world can be found in https://youtu.be/e-7tDhZ4ZgA.

Figures (13)

• Figure 1: Depiction of the reaching task performed by impact hammers in mining, using a Bobcat E10 mini-excavator. To perform this task, the hydraulic arm of the machine has to be controlled so its end-effector reaches target poses from arbitrary initial configurations, using minimal sensing and a discrete control interface at the joint level.
• Figure 2: Simplified electro-hydraulic diagram for the modified Bobcat E10's mini-excavator.
• Figure 3: Overview of the methodology used to obtain reaching controllers for hydraulic impact hammers, given a learned dynamic model $f_{\bm{\theta}}$. Note that estimating $\dot{\bm{q}}_t$ is only necessary when $f_{\bm{\theta}}$ is trained to predict velocity residuals (see Section \ref{['sec:dynamic_model']}).
• Figure 4: Kinematic layout of the Bobcat E10 mini-excavator, alongside a scaled steel grill. The fixed frames $\{B\}$ and $\{G\}$ are attached to the mini-excavator base link and to the center of the grill, respectively. The cylindrical sector defines the boundaries for the position of end-effector poses that may be elements of the restricted workspace $\mathcal{W}^+$.
• Figure 6: Sampled end-effector poses used to set episodic conditions for the reaching task during (a) policy learning, and (b) policy evaluation.