14 OPINION
Lenzo
Dynamics expert, Basilio Lenzo, explains why vehicle state estimation
is a critical factor for reducing road traffi c fatalities, and how it will
help solve the World Health Organization’s concerns for 2030
›Driving a vehicle – or
just being a passenger – is
a very common experience in
our everyday life. With around 1.5 billion
vehicles in the world, the safety of all road
users is of paramount importance. It sounds quite
surprising then that when we drive our cars, some
of the most important parameters related to vehicle
motion – denoted as ‘states’ – are not known.
This lack of knowledge restricts the potential of active
safety systems. ABS and ESC active safety systems are installed
in most modern cars, and they help save thousands of lives
every year, which is remarkable. But many more people still die
in road crashes: with 1.2 million deaths per year, traffi c-related
fatalities are the eighth-leading cause of death worldwide.
The issue is that many key states are very diffi cult – or
even impossible – to measure. For example, a vehicle’s lateral
velocity (alternatively, the side-slip angle) can be measured
using a dedicated sensor, but that has a prohibitive cost
for mass-production vehicles. Tyre-road forces can also be
measured, but practical implications as well as unfeasible
costs mean that this option is not viable. No direct way exists
to measure the tyre-road friction coeffi cient, which depends
on several factors such as road surface, tyre compound,
temperature, etc.
These factors have motivated decades of research into the
development of techniques to estimate relevant vehicle states.
Most of the existing techniques cleverly exploit common
sensory equipment, such as wheel-speed sensors, IMUs (inertial
measurement units) and steering-wheel angle sensors. Even
so, the intrinsic complexity of the studied phenomena has
encouraged researchers to adopt combinations of simplifying
hypotheses, including relatively simple estimation techniques,
simplifi ed vehicle models, and simplifi ed road scenarios (such
as a fl at road), which result in somewhat limited applicability
of the proposed techniques.
In an ideal scenario, a reliable estimator would provide us
with all the relevant vehicle states in any driving condition. In
the short term, not only could ABS and ESC be improved, but
completely new advanced vehicle controllers could be devised.
These controllers would anticipate rather than react to loss of
Basilio Lenzo is a
mechanical engineer
with industrial and
academic experience in
the automotive sector.
Since 2016 he has been
a senior lecturer in
automotive engineering
at Shef eld Hallam
University in the UK,
where he teaches
vehicle dynamics.
His research focuses
on vehicle control
and vehicle state
estimation. You can
watch his last TEDx talk
at https://u.nu/enzo
VehicleDynamicsInternational.com • May/June 2020
control
situations,
which
would be a
paradigm shift
with respect to
existing systems
such as ABS and
ESC. The knowledge of
vehicle states is also one
of the prerequisites for
the large-scale deployment
of autonomous vehicles;
indeed statistics indicate that
more than 90% of accidents are
caused by human error.
Vehicle state estimation can
help to conspicuously improve our
safety, in both the short and long
term. The World Health Organization
suggests that road traffi c fatalities
will become the “fi fth leading cause
of death by the year 2030 unless
urgent action is taken”. Thankfully
vehicle dynamicists and control
engineers are working to develop
the necessary solutions.
Adobe Stock -metamorworks
“The WHO suggests that road traffi c fatalities will become the
“fi fth leading cause of death by 2030 unless urgent action is taken”
/enzo
/VehicleDynamicsInternational.com