1) The document reviews an artificial intelligence microscopic traffic model (AITM) that uses equations of linear and radial motion as well as oriented bounding boxes for collision detection to simulate vehicle behavior.
2) The AITM allows researchers to generate traffic scenarios and optimize traffic control using complex data processing. It provides a virtual environment to study topics like traffic flow, emissions and fuel efficiency.
3) While the AITM provides realistic traffic simulations, it also has limitations like high costs and lack of responsive human-driven vehicles in driving simulators. Future research areas include improved human behavior models and high-fidelity multi-user capabilities.
9953056974 Call Girls In South Ex, Escorts (Delhi) NCR.pdf
AI Traffic Model Review on Microscopic Models Using Oriented Bounding Box
1. BAYERO UNIVERSITY KANO
DEPARTMENT OF CIVIL ENGINEERING
ADVANCED TRAFFIC ENGINEERING
REVIEW ON MICROSCOPIC MODELS
USING ARTIFICIAL INTELLIGENCE {AI}
BY
ANIKOH ABRAHAM OMEIZA
SPS/20/MCE/00017
SUPERVISED BY
PROF. HASHIM M. ALHASSAN
2. INTRODUCTION
• The Artificial Intelligence Traffic Model (AITM) generates a fleet of semi-
intelligent vehicles with which a human driver interacts within a virtual
driving simulation environment. The behavior of the vehicles is based upon
the basic principles of rigid body physics and real-time collision detection,
and includes a rule-base for: road-appropriate travel speed behavior, behavior
at intersections (e.g., stop signs, street lights), and interactions with other AI
and human-driven vehicles on the virtual roads (i.e., lane changing, headway
distance).
3. STATEMENT OF PROBLEM
• Artificial Intelligence in Microscopic Traffic Model is one that is
very difficult to comprehend and its not easily understood and
explained. It is complex and requires critical thinking to
expatiate.
4. AIM OF AI MICROSCOPIC MODEL
• To properly understand the basis of AI in Microscopic traffic
model
5. OBJECTIVES OF AI MICROSCOPIC MODEL
• To clearly understand the concept of AI microscopic model
using Newton’s law of linear and radial motion
• To understand the collision theory of self driving cars using
Oriented Bounding Box
• To be able to clearly and politely criticize present works on AI
Microscopic Model and to give recommendations for future
research.
6. FEATURES OF AI MICROSCOPIC TRAFFIC
MODELS
AI microscopic traffic models are able to process high data travel
information of Drivers/Vehicle.
AI microscopic traffic models simulates increasingly complex scenarios,
helping algorithms learn how to predict vehicular changes.
AI microscopic traffic models provides vehicle–infrastructure integration
(VII) in which vehicles and infrastructure units will communicate with one
another, provides an opportunity to improve the effectiveness and efficiency
of the existing traffic surveillance system.
7. GENESIS AND BASIS OF AITM
• Artificial intelligence (AI) is in the spotlight as one of the emerging fields transforming the transport
sector. It is not a new term. Academics talked about artificial intelligence as early as the 1950s.
• In 1925, inventor Francis Houdina demonstrated a radio-controlled car through the streets of
Manhattan without anyone at the steering wheel. The radio was able to start its engine, shift gears, and
sound its horn. This car offered a glimpse into the future of autonomy but was quickly shut down when
the operator lost control twice during the ride and crashed into another vehicle.
• At the 1939 World’s Fair, General Motors created the first self-driving car model. It was an electric
vehicle guided by radio-controlled electromagnetic fields and operated from magnetized metal spikes
embedded in the roadway.
8. GENESIS AND BASIS OF AITM
• To allow the AI vehicles to navigate within the virtual
environment, a function was designed to dictate their basic
motion path. For this purpose, equations of motion were
implemented based on physics. The motion of the AI vehicles
was decomposed into three broad phases: acceleration,
deceleration, and turning (Newton, 1666).
• The equations of motion are summarized in table below
9. GENESIS AND BASIS OF AITM
Linear Motion Equations Radial Motion Equations
v=s/t
v=v0+a t
s=v0t+1/2a t2
v2=v0
2+2a s
Where:
v= velocity (Km/hr)
s= linear displacement in (Km)
t= time (s)
V0= initial linear velocity (Km/hr)
a= acceleration (Km/hr2)
ω=ɵ/t
ω = ω0 + α t; ω = (ω0
2 + 2 α θ)1/2
θ = ω0 t + 1/2 α t2
α = dω / dt = d2θ / dt2
Where:
ω= angular velocity (rad/s)
θ = angular displacement (rad)
t = time (s)
ɷ0 = angular velocity at time zero (rad/s)
α = angular acceleration (rad/s2)
10. GENESIS AND BASIS OF AITM
• The Linear Motion Equations account for the acceleration and deceleration
motions of the AI vehicles on a straight road. Each time slice was set to the
current frame rate of the simulation, which was approximately 60 Hz.
Maximum travel speeds are calculated dynamically, based largely on the
posted speed limit on each road of travel.
• Regarding turning (radial) motion, outlined in the Radial Motion Equations in
the table above, semi-circular arcs were employed between the end-points
of the lane centers of two adjacent roads. This way, AI vehicular motion
along the turn would be fluent, and angular motion equations were used to
approximate the motion of the vehicle while traversing the arc segment. The
radius of the arc was identified by using the distance between the end
points of adjacent road segments, and the interior angle between the two.
11. GENESIS AND BASIS OF AITM
• Determining if any two 3D objects intersect – a technique
commonly known as collision detection. A common procedure
known as Oriented Bounding Box is used for collision detection
in AITM (Eberly, 2008) This method rotates the bounding box
with the geometry so that the bounding box represents the
geometry of the object, even when rotated. As shown in Figure
1.1, the bounding box is attached to the AI vehicle, and new
coordinates are calculated (using matrix transformations) each
time the AI vehicle moves or rotates.
13. APPLICATIONS OF AITM
• Artificial intelligence traffic model are used to generate
scenarios, optimize control, and predict network behavior from
an operational standpoint. They can provide scientists and
researchers with an overall picture of a hypothetical traffic
system (Boxill and Yu, 2000)
• AI Microscopic Traffic Models Utilizes Oriented Bounding Box
procedure to detect collision. It is created using a
transformation matrix where the box will be translated. The
bounding box is oriented such that the axes are ordered with
respect to the principal components. (Eberly, 2008)
14. ADVANTAGES OF AI TRAFFIC MODEL (AITM)
• AI microscopic traffic model is realistic enough to be able to produce
spontaneous traffic jam as in real traffic.
• AI microscopic traffic model gives accurate prediction of behavior of traffic
flow.
• AI microscopic traffic model is flexible and can compute complex data in a
short period of time.
• With a reliable integrated simulation capacity, researchers could more
accurately study the anticipated impacts of driving on “green”
(environmental) concerns, such as estimated vehicle mileage efficiency and
predicted tailpipe emissions (Hulme et al, 2010). This will become
increasingly relevant with evolving technologies such as hybrid and electric
vehicles, and particularly autonomous driving (Hou et al., 2014).
15. DISADVANTAGES OF AI TRAFFIC MODEL
(AITM)
• A typical Driving Simulators (DS) allows for the analysis of
driver behavior by immersing human subjects within a virtual
simulation environment and monitoring their reactions.
Unfortunately, a Driving Simulators often lacks traffic
authenticity and transportation network realism. In the majority
of simulators, accompanying traffic is pre-programmed, and
does not respond according to the real-time actions of the
human subject who is operating the human-driven vehicle.
(That and Casas, 2011)
• AI Microscopic Traffic Model is a costly model and it is
financially demanding.
16. CURRENT STATE OF RESEARCH
• (Nakasone et al.,2011) introduced OpenEnergySim: a multi-
user driving simulator that is capable of integrating with a
traffic simulator (X-Roads) using the OpenScience framework.
• (Punzo and Ciuffo, 2011) have emphasized the four main
requirements for appropriately integrated (TS-DS) simulation
models. These are: Accurate Road matching between traffic and
driving simulators; Synchronization of traffic and driving
modules with real time; Consistency of the updating calculation
frequency; and Management of autonomous vehicle
visualization.
17. FUTURE RESEARCH
• Implementation of customized human behavior models to
enhance the traffic mobility
• Provide a high-fidelity, multiple-participant capability to
facilitate research that involves real-time interaction between
human participants. (Prendinger et al., 2014)
18. CONCLUSION
• Roadway safety and sustainability continue to be major public health
concerns, and subsequently, simulators (and other M&S technologies)
continue to become more abundant in a wide variety of Intelligent
Transportation Systems (ITS) research applications (e.g., autonomous
driving, human factors, and rehabilitation). To confront these
problems, standalone simulators are often implemented as an
analysis and decision-making tool. The preliminary development of
the core components (i.e., linear motion, radial motion and collision
detection ) of the AITM was described in detail. Subsequent to pilot
testing, the complexity of the interactions between the various model
components was noted, and therefore required numerous revisions
to the baseline AITM model, which were described and illustrated.
19. RECOMMENDATION
• Based on the review conducted, it is recommended that further
traffic research studies be carried out on proper integration of
Driving Simulators and Traffic Simulators in microscopic traffic
model.