4
Hybrid Equipment
Hybrid equipment is both mobile and stationary (rock
drill’s, etc.). In stationary mode, stationary equipment
often possesses articulating components that present haz-
ards to operators and maintenance personnel. This presents
a hierarchy of context that would select one motion model
for stationary mode and another for mobile operation.
Human-operated Equipment
Human-operated equipment has the kinematic constraints
associated with the machine (service vehicles, water truck,
grader, etc.), and the relative indeterminacy of humans.
The motion model would be the same as autonomous,
but context would add large variances to accommodate
the unknown intent of the driver. The value of this object
would reflect the fact that there is an indispensable human
on board. This, coupled with the large variance, would
automatically cause AASIS-enabled equipment to provide
wide berth and slow operation in the presence of human-
operated equipment.
SUMMARY
In order to sufficiently model multiple object motions
for an MSA system, there needs to be object type specific
motion models and a good understanding of all the con-
texts of motion for each type. Creating a Path Projection
engine for MSA starts with a catalogue of expected types,
and all the things that change their behavior. Each type
will also need a motion model appropriate for its particular
kinematics and a clear understanding of how contexts will
influence the motion. Each motion model must be able
to calculate associated statistical variance as a function of
future time. The probability processing engine will use this
information to evaluate likelihoods and costs of interaction
for each pair of objects identified. Ultimate safety of an
AASIS system is heavily dependent on the quality of data
provided by Perception and the adequacy of modeling cre-
ated in Path Projection.
Autonomy is an expensive and complicated endeavor
that is focused on radical improvements in safety and pro-
ductivity. MSA is a crucial enabling element for the tran-
sition from where the mining industry is now to where
it is going. SMRD/NIOSH (Spokane Mining Research
Division of the National Institute for Occupational Safety
and Health) is actively working to identify sensor suites,
software tools, mathematical models, and other emerging
technologies that will help the industry coalesce around a
common, determinant, robust, flexible MSA framework
that will usher in the next generation of safety and produc-
tivity by incorporating the AASIS architype.
REFERENCES
[1] International Organization for Standards,
ISO17757-2019 Earth-moving machinery and min-
ing Autonomous and semi-autonomous machine system
safety, Geneva Switzerland: ISO, 2019.
[2] Computing Community Consortium -Catalyst,
“Assured Autonomy Path Toward Living With
Autonomous Systems We Can Trust,” Computing
Community Consortium -Catalyst, Washington,
2020.
[3] A. S. Mueller, J. B. Cicchino and D. S. Zuby, “What
humanlike errors do autonomous vehicles need to
avoid to maximize safety?” Journal of Safety Research,
pp. 1–4, 2020.
[4] D.E. Smith and J.M. Starkley, “Effects of Model
Complexity on the Performance of Automated Vehicle
Steering Controllers: Model Development, Validation
and Comparison,” Vehicle System Dynamics, vol. 24,
no. 1995, pp. 163–181, 1995.
Hybrid Equipment
Hybrid equipment is both mobile and stationary (rock
drill’s, etc.). In stationary mode, stationary equipment
often possesses articulating components that present haz-
ards to operators and maintenance personnel. This presents
a hierarchy of context that would select one motion model
for stationary mode and another for mobile operation.
Human-operated Equipment
Human-operated equipment has the kinematic constraints
associated with the machine (service vehicles, water truck,
grader, etc.), and the relative indeterminacy of humans.
The motion model would be the same as autonomous,
but context would add large variances to accommodate
the unknown intent of the driver. The value of this object
would reflect the fact that there is an indispensable human
on board. This, coupled with the large variance, would
automatically cause AASIS-enabled equipment to provide
wide berth and slow operation in the presence of human-
operated equipment.
SUMMARY
In order to sufficiently model multiple object motions
for an MSA system, there needs to be object type specific
motion models and a good understanding of all the con-
texts of motion for each type. Creating a Path Projection
engine for MSA starts with a catalogue of expected types,
and all the things that change their behavior. Each type
will also need a motion model appropriate for its particular
kinematics and a clear understanding of how contexts will
influence the motion. Each motion model must be able
to calculate associated statistical variance as a function of
future time. The probability processing engine will use this
information to evaluate likelihoods and costs of interaction
for each pair of objects identified. Ultimate safety of an
AASIS system is heavily dependent on the quality of data
provided by Perception and the adequacy of modeling cre-
ated in Path Projection.
Autonomy is an expensive and complicated endeavor
that is focused on radical improvements in safety and pro-
ductivity. MSA is a crucial enabling element for the tran-
sition from where the mining industry is now to where
it is going. SMRD/NIOSH (Spokane Mining Research
Division of the National Institute for Occupational Safety
and Health) is actively working to identify sensor suites,
software tools, mathematical models, and other emerging
technologies that will help the industry coalesce around a
common, determinant, robust, flexible MSA framework
that will usher in the next generation of safety and produc-
tivity by incorporating the AASIS architype.
REFERENCES
[1] International Organization for Standards,
ISO17757-2019 Earth-moving machinery and min-
ing Autonomous and semi-autonomous machine system
safety, Geneva Switzerland: ISO, 2019.
[2] Computing Community Consortium -Catalyst,
“Assured Autonomy Path Toward Living With
Autonomous Systems We Can Trust,” Computing
Community Consortium -Catalyst, Washington,
2020.
[3] A. S. Mueller, J. B. Cicchino and D. S. Zuby, “What
humanlike errors do autonomous vehicles need to
avoid to maximize safety?” Journal of Safety Research,
pp. 1–4, 2020.
[4] D.E. Smith and J.M. Starkley, “Effects of Model
Complexity on the Performance of Automated Vehicle
Steering Controllers: Model Development, Validation
and Comparison,” Vehicle System Dynamics, vol. 24,
no. 1995, pp. 163–181, 1995.