Rope Shovel Automation

This project has demonstrated a number of technologies that are critical to automating rope shovel operation. They are:

• Computer control of all three drives of a model rope shovel (crowd, hoist and swing).
• Reliably detecting the onset of stall and two strategies to modify the digging motion to avoid the stall condition.
• Creation of 3D terrain maps using a laser scanner which provides information for automatically engaging the bank, as well as providing a situation display for the operator.
• Two approaches to determining when the dipper is full have been investigated and show considerable promise for automatically determining when to disengage the dipper from the bank. The methods are:
• based on measured motor signals such as current and speed
• based on a laser scanner observing a cross-section across the dipper

This report presents observations on the operation of a production rope shovel and a review of other work in the area of shovel automation and dipper weight estimation. Results on laser DTM generation, machine automation, and digging experiments are then presented. The experimental program was conducted using a 1/7th scale model rope shovel based at Redbank in Brisbane.

Based on these encouraging results it would be appropriate to consider the next phase of research which could take one of two courses:

1. Continue working with the model shovel and demonstrate complete cycles and dumping into a model sized truck tray.
2. Commission the automation system on a production rope shovel.

The similarity between rope shovels and draglines, and previous experience with dragline automation, suggests that the second approach would be quite feasible.