Digging for Data: Experiments in Rock Pile Characterization Using Only Proprioceptive Sensing in Excavation

TL;DR

This work demonstrates a wavelet-based framework for estimating relative rock pile fragmentation using only proprioceptive sensing from excavation equipment, avoiding exteroceptive sensors. By computing a wavelet-derived feature, $\zeta$, from bucket and boom signals, the authors show that pile-to-pile ratios approximate relative mean particle sizes and align with ground-truth references from sieve analysis and a vision-based tool. Field experiments with full-scale, battery-electric LHDs across five rock piles validate the method and reveal that operator behavior can influence results, underscoring the method's suitability for autonomous excavation where consistent control policies minimize such variability. The study advances autonomous material handling by enabling rapid, bucket-level fragmentation assessment in harsh, dusty environments, potentially informing adaptive control and classification tasks in real time.

Abstract

Characterization of fragmented rock piles is a fundamental task in the mining and quarrying industries, where rock is fragmented by blasting, transported using wheel loaders, and then sent for further processing. This field report studies a novel method for estimating the relative particle size of fragmented rock piles from only proprioceptive data collected while digging with a wheel loader. Rather than employ exteroceptive sensors (e.g., cameras or LiDAR sensors) to estimate rock particle sizes, the studied method infers rock fragmentation from an excavator's inertial response during excavation. This paper expands on research that postulated the use of wavelet analysis to construct a unique feature that is proportional to the level of rock fragmentation. We demonstrate through extensive field experiments that the ratio of wavelet features, constructed from data obtained by excavating in different rock piles with different size distributions, approximates the ratio of the mean particle size of the two rock piles. Full-scale excavation experiments were performed with a battery electric, 18-tonne capacity, load-haul-dump (LHD) machine in representative conditions in an operating quarry. The relative particle size estimates generated with the proposed sensing methodology are compared with those obtained from both a vision-based fragmentation analysis tool and from sieving of sampled materials.

Figures (18)

• Figure 1: Example excavation operation where the battery electric load-haul-dump (LHD) used in the study presented by this field report excavates fragmented rock into its bucket at an operating quarry.
• Figure 2: Example images at different stages of the excavation process and the variability of the visible rocks. A ruler of length 1.2 m is provided for scale and is roughly the same number of pixels in each image.
• Figure 3: Example workflow of the vision based fragmentation analysis tool WipFrag™ and resulting fragmentation report.
• Figure 4: Example pile images of the four semi-homogeneous aggregate piles used for excavation experiments. Two 1.2-m wide rulers with 2-cm yellow squares spaced 0.4 m apart is provided for scale.
• Figure 5: Pile images of the heterogeneous aggregate pile 0/1500. Two 1.2-m wide rulers with 2-cm yellow squares spaced 0.4 m apart is provided for scale.