Smart snake reconnaissance system [SASAR]

Abstract

The motivation for this thesis is to develop a new, innovative snake-like robot platform: the Smart Snake Reconnaissance System (SASAR). The snake is chosen as the basis of our design due to their natural adaptability and the versatility of environments they are able to inhabit and traverse through. We therefore design a robot capable of a wide range of applications, from maintenance, to exploration, to photography, in both defence and commercial scenarios. This PhD research has especially focused on three core design pillars: Modularity, Semi-Autonomy, and Novelty. Four prototype joints to actuate the new robot were designed, following an investigation which examined and assessed existing snake-like robots, and other innovative robotic mechanisms. We analysed their physical characteristics and investigated how the number of degrees of freedom within the mechanism affect their performance. Additionally, research and development were conducted into different methods of control for a snake-like robot. This led to the creation of both 2D and 3D simulations to analyse the motion of the new robot design. Finally, we present our new snake-like robot, consisting of four independent, low cost, 3D printed modules. Our design implements modularity through the use of a common electrical and mechanical connector, plus we utilise a CAN bus to communicate between the modules of the robot. Semi-autonomy has been implemented by developing a kinematic model for the control of the robot, which takes two simple inputs and calculates the motion required for each link in the chain. The novelty mainly reflects in the design and manufacture of the unique “Quaternion” style Joint module, which uses a complex three-bar parallel linkage to actuate the spine of the robot. The performance of this Joint Module is then analysed, compared to previous snake-robot designs. We therefore manage to achieve our initial goal of creating a novel, modular, snake-like robot.