1
24-033
Enhancing Robotic Perception for Autonomous Roof Bolting
Using an Event Based Machine Learning Framework
Akram Marseet
M3 Robotics Lab
Rik Banerjee
M3 Robotics Lab
Andrew J. Petruska
M3 Robotics Lab
ABSTRACT
Underground mine roof bolting is a crucial operation for
miners’ safety and mine sustainability. Since roof bolting is
a manual or human-supervised operation, miners’ safety is
at risk due to dust or rock falls. Traditional machine learn-
ing algorithms have shown limitations to detecting drillable
areas, mainly due to harsh lighting conditions. The authors
propose an adaptive deep-learning framework for autono-
mous roof bolting. The proposed framework is based on
implementing a binary semantic segmentation algorithm
on color images to classify pixels that belong to rock from
those that belong to non-rock. Significantly, the proposed
framework implements deep learning semantic segmenta-
tion on images from traditional and neuromorphic vision
sensors in underground mines. The performance of the
proposed model shows an impressive accuracy level of at
least 98% at a low number of training epochs with smooth
learning curves. The high accuracy enables the implementa-
tion of autonomous roof bolting, greatly improving min-
ers’ safety and operational efficiency while reducing human
exposure to safety hazards. This research will advance the
use of deep learning in mining automation and has the
potential to revolutionize the traditional mining industry.
INTRODUCTION
The Mining sector plays a pivotal role in the develop-
ment of the world economy [1], [2]. Underground min-
ing is growing due to the rising demand for minerals for
industries such as the automotive industry and clean energy
technologies [3]. Due to the rapid demand for minerals,
This work is supported by National Institute for Occupational
Safety &Health |NIOSH/Project 75D30121C12206.
the mining industry is moving toward deep underground
mining [4]. Nonetheless, underground mining presents sig-
nificant challenges across various dimensions [5]. Human
safety is one of the main challenges in underground mining
[6], mines sustainability [7], and working conditions [8]
also contribute to other challenges in underground mining
environments. According to a report issued by the United
States Mine Safety and Health Administration (MSHA),
the total number of fatal injuries till the third quarter of
2023 is higher than that for the previous year [9].
One of the hazards in the underground mining industry
that causes fantails and serious injuries is falls on the roofs
and the ribs of the underground mines. [10]. According to
MSHA, 26 fatal injuries have been reported as of August
5th, 2023 [11]. Roof fall hazards are addressed by roof
bolting [12] which has been recognized by the Coal Mine
Health and Safety Act of 1969 as the exclusive method of
providing support for underground entry [13]. Roof bolt-
ing is a fundamental technique for ensuring the safety and
stability of underground mine environments, particularly
in areas where roof conditions are unstable or prone to
collapses [14]. Roof bolting is still an active research area
where there are many aspects that need to be addressed
including safety during roof bolting and increasing the effi-
ciency of the roof bolting process. Roof bolting processes
mainly include perforating the unsupported roof, followed
by the insertion of a roof bolt and epoxy resin to fasten the
overlying roof strata [15]. For rock support, straps are used
[16]. Therefore, an additional step is added which is drilling
through holes in the straps.
Although roof bolting improves the safety of mine
personnel and the sustainability of mines, the roof bolting
process is risky on roof blotters. There are safety hazards
24-033
Enhancing Robotic Perception for Autonomous Roof Bolting
Using an Event Based Machine Learning Framework
Akram Marseet
M3 Robotics Lab
Rik Banerjee
M3 Robotics Lab
Andrew J. Petruska
M3 Robotics Lab
ABSTRACT
Underground mine roof bolting is a crucial operation for
miners’ safety and mine sustainability. Since roof bolting is
a manual or human-supervised operation, miners’ safety is
at risk due to dust or rock falls. Traditional machine learn-
ing algorithms have shown limitations to detecting drillable
areas, mainly due to harsh lighting conditions. The authors
propose an adaptive deep-learning framework for autono-
mous roof bolting. The proposed framework is based on
implementing a binary semantic segmentation algorithm
on color images to classify pixels that belong to rock from
those that belong to non-rock. Significantly, the proposed
framework implements deep learning semantic segmenta-
tion on images from traditional and neuromorphic vision
sensors in underground mines. The performance of the
proposed model shows an impressive accuracy level of at
least 98% at a low number of training epochs with smooth
learning curves. The high accuracy enables the implementa-
tion of autonomous roof bolting, greatly improving min-
ers’ safety and operational efficiency while reducing human
exposure to safety hazards. This research will advance the
use of deep learning in mining automation and has the
potential to revolutionize the traditional mining industry.
INTRODUCTION
The Mining sector plays a pivotal role in the develop-
ment of the world economy [1], [2]. Underground min-
ing is growing due to the rising demand for minerals for
industries such as the automotive industry and clean energy
technologies [3]. Due to the rapid demand for minerals,
This work is supported by National Institute for Occupational
Safety &Health |NIOSH/Project 75D30121C12206.
the mining industry is moving toward deep underground
mining [4]. Nonetheless, underground mining presents sig-
nificant challenges across various dimensions [5]. Human
safety is one of the main challenges in underground mining
[6], mines sustainability [7], and working conditions [8]
also contribute to other challenges in underground mining
environments. According to a report issued by the United
States Mine Safety and Health Administration (MSHA),
the total number of fatal injuries till the third quarter of
2023 is higher than that for the previous year [9].
One of the hazards in the underground mining industry
that causes fantails and serious injuries is falls on the roofs
and the ribs of the underground mines. [10]. According to
MSHA, 26 fatal injuries have been reported as of August
5th, 2023 [11]. Roof fall hazards are addressed by roof
bolting [12] which has been recognized by the Coal Mine
Health and Safety Act of 1969 as the exclusive method of
providing support for underground entry [13]. Roof bolt-
ing is a fundamental technique for ensuring the safety and
stability of underground mine environments, particularly
in areas where roof conditions are unstable or prone to
collapses [14]. Roof bolting is still an active research area
where there are many aspects that need to be addressed
including safety during roof bolting and increasing the effi-
ciency of the roof bolting process. Roof bolting processes
mainly include perforating the unsupported roof, followed
by the insertion of a roof bolt and epoxy resin to fasten the
overlying roof strata [15]. For rock support, straps are used
[16]. Therefore, an additional step is added which is drilling
through holes in the straps.
Although roof bolting improves the safety of mine
personnel and the sustainability of mines, the roof bolting
process is risky on roof blotters. There are safety hazards