Design

The design criteria our project must meet include:

Sensing: using LiDAR to sense the maze walls and onboard RealSense camera to identify any obstacles in view
Planning: implementing code that will be able to figure out movements and steps to take once the robot reaches a dead end
Actuation: creating an actuator that can pick up objects from the ground

The desired functionality is:

Enhancing the maze's adjustability and reproducibility for testing purposes
Navigate the maze without getting too close to the walls
Turning once the robots sees an AR tag
Continuously move throughout the maze without bumping into the walls with minimal guidance

The design components we chose includes 3-D printed maze posts, a reconfigurable maze design, 3-D printed actuator with 3 motors that can pick up objects, and code that includes AR detection and PID control logic.

Maze Posts

Design: The custom maze posts were 3-D printed.

Design Choices: They have right angle joints so that custom mazes can be created and adjusted with ease. They also have space at the bottom to hold up to two 50 gram weights.

Trade-Offs: The material is light and easy to carry, but some of them cracked near the weight insert area after repeated use.

Real Engineering Applications: The posts meet real-world criteria as they are efficient and easy to use.

Maze Design

Design: The maze is built using A4 size corrugated cardboard as walls.

Design Choices: It is adjustable and can be designed in any way.

Trade-Offs: Since cardboard is used for the walls, the cardboard can get flimsier as it is being used. Additionally, if the robot accidentally bumps into the walls, the maze needs to be partially rebuilt.

Real Engineering Applications: This maze meets engineering criteria as it is fairly robust and durable.

Actuator Design

Design: The actuator moves up and down and closes once it is in front of the object so that it can pick it up.

Design Choices: The actuator includes three motors that drive the vertical movement.

Trade-Offs: It was very difficult to get the right measurements of the actuator components. There was a lot of re-printing and measuring required.

Real Engineering Applications: The actuator meets industry standards as it can be utilized to pick up objects. The actuator is also quick to move up and down.

Code Design

Design: The code design includes AR detection and PID control logic.

Design Choices: The speed of the robot will slow down as it gets closer to the AR tags.

Trade-Offs: The code requires testing to figure out whether or not the RealSense camera can detect AR tags as it is rotating left or right.

Real Engineering Applications: The code's engineering applications encompass obstacle avoidance and minimizing potential damage which is important for real-world robotics in terms of safety.

Overall, the design choices that we utilized demonstrate a comprehensive approach to address the desired criteria and functionality. The use of LiDAR and a RealSense camera for sensing, along with the implementation of code highlights a strategy for maze navigation.

Page updated

Report abuse