Soft Computing Techniques in Electrical Engineering (Ee571) : Topic-Lateral Control For Autonomous Wheeled Vehicle

SOFT COMPUTING TECHNIQUES IN
ELECTRICAL ENGINEERING(EE571)
TOPIC-LATERAL CONTROL FOR
AUTONOMOUS WHEELED VEHICLE
Professor In-charge: Dr D.S Acharya
Name: Shivam Singh
Roll No: MT/EE/10005/23

VEHICLE MODELLING
The kinematic model, as its name
suggests, focuses on the vehicle's
motion without considering the forces
acting on it. It describes the vehicle's
movement in terms of position,
velocity, and acceleration, making it
more sophisticated than simple
geometric models. However, it
operates under specific assumptions
and is most effective when there are no
slip angles, meaning the velocity
vector aligns with the wheels'
orientation. This condition holds true at
low speeds, typically below 5 m/s.
The transition from a kinematic model
to a dynamic model for a vehicle,
emphasizing the need for a more
accurate representation that considers
dynamic behaviors such as side-
slipping, oversteering, and friction. The
dynamic model incorporates forces
acting on the vehicle, especially tire
forces, using Newton's second law and
Euler Lagrange methods. The complete
dynamic model is complex and
nonlinear, addressing translation and
rotation motions in 3D space for a full
four-wheeled vehicle.
Although the single-track bicycle
dynamic model is fit for control design,
it lacks precision due to the
simplifications and assumptions to build
the model. As discussed earlier, full
vehicle dynamic models are a more
realistic representation of the vehicle
dynamics. Such models take into account
all four wheels of the vehicle and do not
neglect the vertical dynamics as
presented in earlier.
VEHICLE CONTROL
• Automatic control is an essential task for autonomous driving systems; it solves the problem of following a
set reference defined as a trajectory or a speed profile depending on the type of control. Control is a complex
task for autonomous vehicles because it must ensure the stability of the vehicle and certain levels of
performance. In general, the control task accounts for Two aspects:
• The type of control which can be lateral control, longitudinal control or both at the same time.
• The vehicle model used for implementing the control which can be kinematic or dynamic and linear or non-
linear.
MODEL BASED CONTROLLER
1)Model Predictive Control(MPC): MPC is a
well-established strategy known as receding
horizon control. MPC is based on the
formulation of an optimization problem in the
form of a finite horizon open-loop optimal
control problem. The main idea of MPC is in
using the discrete model of the system for
predicting its behaviour into the future up to a
certain time step called prediction horizon. A
control input sequence is generated along
several discrete time steps known as the control
horizon. The control sequence solves the
optimization problem by minimizing a cost
function while obeying to certain constraints. At
each time step, a finite horizon window
(prediction horizon) is shifted, and new
measurements of the system and the
environment are used to solve the optimization
problem resulting in a new control sequence.
• 2)Sliding Mode Controller: In sliding
mode control, a sliding surface is a concept
used to describe a specific state subspace in
the system's state space. The sliding surface
plays a crucial role in sliding mode control
as it is designed to guide the system's state
trajectories to remain on or converge to this
surface during the control process. The idea
behind sliding mode control is to force the
system to reach and stay on the sliding
surface, where a specific control law is
applied. When the system is on the sliding
surface, it is considered to be in a "sliding
mode," and the control law keeps the system
dynamics on this surface, leading to
desirable performance and stability
characteristics.
3) H∞ controller: The H∞ control method is a
robust technique designed to handle uncertainties
and external disturbances in a system. It aims to
minimize the H∞ norm through an optimization
process involving the Riccati equation. This
approach, known for its robust stability and
performance, tackles control problems by
considering noise, modelling uncertainties, and
disturbances, often employing Linear Matrix
Inequality (LMI) methods for solution. Despite
its effectiveness, implementing H∞ control can
be challenging in practice. To apply H∞ control
to lateral vehicle dynamics, a system model is
utilized, including equations defining vehicle
dynamics and accounting for external
disturbances. For a more detailed exploration of
H∞ control theory and design.
4)LQR Controller: The LQR approach is
an optimal control technique that achieves
great results at minimum cost. The idea is
similar to MPC where a quadratic cost
function is minimized resulting in an
optimal feedback gain; this is achieved by
solving the Riccati equation which is
formulated based on the plant's state space
model. In lateral control applications,
designing the LQR controller is a straight
forward task. Considering the vehicle
dynamics model, an error-based dynamic
model can be deduced. Assuming that e1 is
the lateral error (between trajectory and
vehicle lateral position) and e2 is the
heading error (between vehicle and
reference heading angles), the following
relations hold true:
ATPA−(ATPB)(R+BTPB)−1(BTPA)+Q=0
CONTROLLER BASED ON ARTIFICIAL
INTELLIGENCE
1)Supervised Learning Control
Fully connected networks are composed of

multiple layers of neurons interconnected by
synaptic weights; they are inspired from brain
neural networks. Generally, they consist of one
input layer, one output layer, and one or more
hidden layers which define the depth of the
network. The training process of neural
networks is mostly based on the gradient descent
algorithm. The forward pass propagates the
inputs throughout the network layers, where
each neuron normalizes its output by an
activation function like the Sigmoid.
2)Reinforced Learning Control
Reinforcement learning is an unsupervised approach where

the network learns from trial and error, this is usually
modelled as a Markov decision process (MDP) illustrated . In
a general way, an MDP is defined by a state space S, an action
space A, a state transition probability P, and a reward function
R. The goal is to learn a policy 𝜋(st, at) that links rewards and
actions such that at each iteration, the agent observes the set
of states st and takes an action at from the action space A.
Then the environment transitions according to probability
model P.
3)Fuzzy Logic Control

Fuzzy logic systems seek to model human expertise by using
linguistic variables; these variables conceptualize the fuzziness
in human decision-making. Such systems do not require high
computational costs unlike conventional control methods, and
they do not need any mathematical model which makes them
suitable for systems with complex models. Human operational
experience is then conceptualized into a fuzzy controller that
consists of (IF/THEN) rules.
CONCLUSION AND FUTURE SCOPE
• In conclusion, the paper presents a comparative analysis of various control methods, including model-based
and model-free approaches, LQR-based lateral control, sliding mode control, H∞ control, and model
predictive control. The paper also highlights the importance of considering constraints and safety conditions
in control design, as well as the need to address faulty sensors and sensing limits.
• As for future scope, the paper suggests that vehicle steering systems should be considered more in control
design, and that data-based methods and black-box techniques may become more prevalent. The paper also
suggests that stability analysis for model predictive control remains a challenge, and that H∞, SMC, and
back-stepping control may be useful for dealing with nonlinearities, parameter uncertainties, and
disturbances. Finally, the paper suggests that the combination of model-based and model-free control
methods may provide a better overall solution in many situations.
REFERENCES
1. NHTSA: Fatal motor vehicle crashes: Overview. U.S. Departement of Transportation, 2019.
https://crashstats.nhtsa.dot.gov/ Api/Public/ViewPublication/812826
2. R. Behringer, S. Sundareswaran, B. Gregory, R. Elsley, B. Addison, W. Guthmiller, R. Daily, and D. Bevly, The
DARPA grand challenge—Development of an autonomous vehicle, Ieee intelligent vehicles symposium, 2004, 2004,
pp. 226–231.
3. P. Sun, H. Kretzschmar, X. Dotiwalla, A. Chouard, V. Patnaik, P. Tsui, J. Guo, Y. Zhou, Y. Chai, and B. Caine,
Scalability in perception for autonomous driving: Waymo open dataset, Proceedings of the ieee/cvf conference on
computer vision and pattern recognition, 2020, pp. 2446–2454.
4. B. Paden, M. Cáp, S. Z. Yong, D. Yershov, and E. Frazzoli, ˇ A survey of motion planning and control techniques for
self-driving urban vehicles, IEEE Trans. Intell. Veh. 1 (2016), no. 1, 33–55.
5. O. Amidi and C. E. Thorpe, Integrated mobile robot control, Mobile Robots V 1388 (1991), no. May, 504–523.

Soft Computing Techniques in Electrical Engineering (Ee571) : Topic-Lateral Control For Autonomous Wheeled Vehicle

Uploaded by

Copyright:

Available Formats

Soft Computing Techniques in Electrical Engineering (Ee571) : Topic-Lateral Control For Autonomous Wheeled Vehicle

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Soft Computing Techniques in Electrical Engineering (Ee571) : Topic-Lateral Control For Autonomous Wheeled Vehicle

Uploaded by

Copyright:

Available Formats

SOFT COMPUTING TECHNIQUES IN

TOPIC-LATERAL CONTROL FOR

AUTONOMOUS WHEELED VEHICLE

Professor In-charge: Dr D.S Acharya

Name: Shivam Singh

Roll No: MT/EE/10005/23

Fully connected networks are composed of

Reinforcement learning is an unsupervised approach where

3)Fuzzy Logic Control

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.