High Fidelity IMU and Wheel Encoder Models for ROS Based AGV Simulations

In this research chapter, we study the problem of how automated guided vehicles (AGVs) are simulated considering the realistic settings in their sensors. Wheel encoders and inertial measurement units (IMUs) are two fundamental sensors in AGVs. These sensors are prone to error due to internal or environmental disturbances and noises. To have a better representation of a robot in simulation environments, these disturbances/noises should also be modeled. However, since simulators have the simplified or idealized version of sensors, the high fidelity models of the sensors are usually ignored, which leads to simulation results inconsistent with the real-life scenarios. In this chapter, we develop related ROS nodes and Gazebo plugins to incorporate the disturbance/noise into the sensors and model the error occurrences using Poisson distribution. We validate the approach through Gazebo simulations using an AGV model. The source codes and installation details are provided via a public repository.