Mobile Robot Control 2023 Rosey
Welcome to the group page of team Rosey!
Group members table
| Name | Student ID |
|---|---|
| Eline Wisse | 1335162 |
| Lotte Rassaerts | 1330004 |
| Marijn Minkenberg | 1357751 |
| Rainier Heijne | 1227575 |
| Tom Minten | 1372300 |
| Lowe Blom | 1266020 |
Midterm presentation
File:Midterm presentation Rosey.pdf
Current state diagram
File structure
Work division
We identified three 'major parts' of the assignment. These are the localisation, the global and local navigation of the robot. We additionally saw use for an overseeing role, which should ensure that the interfaces between the functions are clear. The work is divided as follows:
- Marijn : System architecture & overview
- Tom : Global navigation (A*)
- Eline & Rainier : Local navigation (potential fields + open space)
- Lowe & Lotte : Localisation
Used algorithm and predefined knowledge
The global navigation uses the A* algorithm to find the shortest path to from a starting node to a goal node. The A* algorithm uses a predefined grid of nodes. This grid is applied by a JSON file, generated with Matlab. With Matlab, the grid nodes with their connections are manually chooses in the known map. On top of that, we were inspired by the lecture about Safe navigation in a hospital environment to introduce some context-aware navigation into our model, by means of semantics. Each node contains some semantic information about its environment, described by a integer. The following semantic integers are included: 0 = table 0, 1= table 1, ..., 49 = table 49, 50 = door, 99 = empty (meaning no semantic information present). This information is also included in the JSON file
Initiation phase (run once)
The json file is imported with the loadConfigFile()and the grid is constructed withconstructGrid().