BNU - Faculty of Engineering
Mechatronics Department
Modeling a Two-Wheel
Robot
Prepared by: Zyad Hany
Supervised by: Prof. Dr. Aref Soliman
Contents
1. Introduction ................................................................................................................................ 3
2. System Description...................................................................................................................... 3
Robot Parameters ........................................................................................................................... 3
3. Kinematic Modeling .................................................................................................................... 4
Forward Kinematics .................................................................................................................... 4
Position Update Equations .......................................................................................................... 4
4. Dynamic Modeling ...................................................................................................................... 4
Equations of Motion ................................................................................................................... 5
5. Control Strategies ........................................................................................................................ 5
6. Simulation ................................................................................................................................... 5
7. Applications ................................................................................................................................. 5
8. Conclusion ................................................................................................................................... 6
1. Introduction
Two-wheel robots are among the most fundamental types of mobile robots used in research,
education, and industry. These robots typically operate using differential drive mechanisms,
where each wheel is independently powered. This report presents a basic kinematic and
dynamic model of a two-wheel robot, covering key components and assumptions necessary for
control, simulation, and real-world implementation.
2. System Description
A two-wheel differential drive robot consists of:
• Two drive wheels (left and right), mounted on the same axis.
• Optional caster wheel(s) to support balance (not powered).
• Chassis/body housing the controller, power supply, and sensors.
• Motors to drive each wheel independently.
Robot Parameters
• R: Radius of the wheels
• L: Distance between the two wheels (wheelbase)
• 𝑣𝑙 , 𝑣𝑟 : Linear velocities of the left and right wheels
• 𝜃: Robot's heading angle
• (x, y): Robot’s position in the 2D plane
3. Kinematic Modeling
The robot's motion is governed by non-holonomic constraints (i.e., it cannot move sideways).
Using the differential drive kinematics:
Forward Kinematics
The linear and angular velocity of the robot are calculated as:
𝑣𝑟 + 𝑣𝑙
𝑣=
2
𝑣𝑟 − 𝑣𝑙
𝜔=
𝐿
Position Update Equations
𝑥̇ = 𝑣 cos(θ)
𝑦̇ = 𝑣 sin(θ)
𝜃˙ = 𝜔
These equations allow the prediction of the robot's position and orientation over time.
4. Dynamic Modeling
To develop control algorithms or simulate real-world dynamics, a dynamic model can be used.
Assuming a lumped mass model with wheel torque inputs:
Equations of Motion
Using Newton's laws or Lagrangian methods, we derive:
𝑀𝑥¨ + 𝐶(𝑥˙) + 𝐷 = 𝐹
Where:
• M: Mass/inertia matrix
• C: Coriolis and centrifugal forces
• D: Damping or friction forces
• F: Input force or torque from motors
Each wheel's torque τ𝑙 , τ𝑟 produces a force:
τ
𝐹=
𝑅
This can be extended to model slip, friction, or more complex motor dynamics.
5. Control Strategies
Typical control techniques include:
• PID control for velocity or position
• Trajectory tracking controllers using feedback linearization or Lyapunov methods
• Model Predictive Control (MPC) for optimized motion planning
Sensor feedback (from encoders, IMUs, or cameras) is often used to correct errors in motion.
6. Simulation
Simulation environments like MATLAB/Simulink, Python (with libraries like ROS, Gazebo, or
PyBullet), or CoppeliaSim are used to test kinematic and dynamic models. These platforms
support integration with sensor models and control algorithms.
7. Applications
Two-wheel robots are used in:
• Educational platforms (e.g., TurtleBot, ePuck)
• Indoor navigation and mapping
• Line following and obstacle avoidance
• Research in mobile robotics and AI
8. Conclusion
Modeling a two-wheel robot involves both kinematic and dynamic aspects that are essential for
accurate control and simulation. A solid understanding of the system's parameters, constraints,
and motion equations enables the development of efficient control algorithms and practical
robotic applications.
