From conceptual to actual design of unmanned vehicle

• Main Author: Ng, Alvin Heng San.
• Other Authors: Zhong Zhaowei
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Engineering::Mechanical engineering::Motor vehicles
• Online Access: http://hdl.handle.net/10356/40157
• Institution: Nanyang Technological University

From the conceptual design selection process to the phase two prototype, a miniaturized (170mm x 105mm x 80mm) Unmanned Surveillance Ground Vehicle (USGV) is built from scraps to test its capabilities. Using the findings from the literature review, it is discovered that there is classification of vehicles which gives it different terrain capabilities. Steering and driving mechanism have to be selected to suit the classification of the vehicles. Energy used for propulsion and steering transmission are also considered together with the controller required for the application. These are the factors that set the difference between a manned and an unmanned vehicle. The methodology uses the findings from the literature review, organized into a series of flexible stages to be used in the design selection process to build the miniaturized prototype. The stages act as guidelines, and are useful in the selection process. It consists of the morphological chart selection, conceptual and actual design component and the prototype phase. Conceptual designs were made and the suspension system was omitted due to time and budget constraints. The actual design involves the sourcing of components from various suppliers and finally the building of the prototype. The selected components price list and descriptions are also presented. The prototype went through two phases with operational test conducted. However it has yet to be fully developed. In phase one, the USGV prototype is tele-operated via a remote control by an operator. Three problems are identified; unsynchronized differential-drive motor speed, off-centered Center of Mass, and operator’s blind spot. In phase two, the USGV prototype is semi-autonomous, able to automatically divert from obstacles in the operator’s blind spots. The USGV is still able to receive directional instructions from the operator via the remote controller. The purpose of phase two is to solved the three main problems identified in phase one. 4 The unsynchronized differential-drive motor speed still remain as a main problem as the BASIC Stamp controller used only output a single forward speed for the autonomous mode, and fine tuning by the operator’s remote control trimmer is not allowed as it may upset the program algorithm. The off-centered Center of Mass problem is solved with the additional electronic components acting as ballast weights. Thus the overall obstacle climbing capabilities of the USGV have improved from the previous 15 mm to 20 mm, given the small prototype size. Although the operator’s blind spot have been overcome by using Ping sensors to detect obstacles, the USGV prototype is only semi-autonomous as there are also sensor blind spots. Along with the new capabilities discovered in phase two, such as the ability to avoid obstacles automatically, guide along peripheral walls, “stop” when met with difficult obstacles, opposed operator’s input to avoid collision; there also limitations. Regardless of its capabilities or limitations, there are more positive implications than negative implications. This had proved that the objective of this project, the semi-autonomous control concept to reduce operator workload of obstacle detection and avoidance has been fairly successful. Recommendations are also made for phase three prototype to improve the prototype, either with add-ons to the current prototype or to built a bigger prototype, with the findings from phase one and two.