Simulation on Smart Vehicle Navigation with Multiple Sensors

2021 ELE Engineering Design Project (XZ05)

---

Faculty Lab Coordinator

Xiao-Ping Zhang

Topic Category

Signal Processing / Communication

Preamble

A smart vehicle is equipped with multiple sensors which collects absolute localization information in a specific area, and relative distance from an obstacle, heading angle information, etc. Given the coordinate (may change with time) of the destination and the above data from sensors, the vehicle needs to determine where it moves in the next step.

Objective

Understanding how to generate simulated sensor data based on senor models. Learn principles to make decision for vehicle movement automatically using noisy sensor data. Visualize the simulation scene and show the decision process and moving trajectory of the vehicle in a real-time manner.

Partial Specifications

The simulation software supports configuration on area size and shape, number/size/shape of obstacles, selection of used sensors, errors of the sensor data.
The available simulated sensors must include at least 3 types of sensors including but not limited to position sensors (e.g., GPS or UWB localization), relative attitude or posture sensors (e.g., IMU, magnetometer), absolute attitude or posture sensors (e.g., magnetometer), distance sensors (e.g., ultra-sound, mmWave radar).
Use GUI for user input and output display.

Suggested Approach

Learn from textbooks and technical webpages to understand the error models of the sensors. Generate sensor data according to user input error scales as input to the navigation algorithm for vehicle. The navigation algorithm needs to make decisions on how the vehicle will move in order to avoid the obstacles reach the destination in a short path and short time. The GUI receives commands from the user, display the process of the vehicle movement and decision, and store the data throughout the process.

Group Responsibilities

Coordinate to define software interfaces and to divide tasks. Realize and demonstrate the entire software system. Deliver the design documents for each module and interfaces.

Student A Responsibilities

Program a software module for real-time sensor data generation. Learn the features from practical low cost sensors as a foundation. Need to define data interface with the GUI and the navigation algorithm. For obstacle detection, need to coordinate with GUI to define different obstacles and corresponding sensor data.

Student B Responsibilities

Navigation algorithm realization. Explore different methods for decision making based on external sensor data. Define input and output data structure and protocols with GUI and data generation modules. Discuss with Student C on how to divide task and realize this module. Can have different decision making algorithm from Student C.

Student C Responsibilities

Navigation algorithm realization. Explore different methods for decision making based on external sensor data. Define input and output data structure and protocols with GUI and data generation modules. Discuss with Student B on how to divide task and realize this module. Can have different decision making algorithm from Student B.

Student D Responsibilities

Realize GUI for user input and data visualization. Show the trajectory of the movement of the vehicle in a real-time manner. Receive user input for area size/shape, origin/destination coordinates, size/shape of obstacles, start/stop of vehicle movement, etc.

Course Co-requisites

To ALL EDP Students

Due to COVID-19 pandemic, in the event University is not open for in-class/in-lab activities during the Winter term, your EDP topic specifications, requirements, implementations, and assessment methods will be adjusted by your FLCs at their discretion.