Gait identification modelling for a hybrid amphi-underwater robot with exponential momentum ant colony algorithm

• Main Authors: Toha, Siti Fauziah, Ahmad, Salmiah, Ismail, Amelia Ritahani, Abdul Hamid, Syamsul Bahrin
• Format: Monograph
• Language: English
• Published: 2020
• Subjects: T Technology (General)
• Online Access: http://irep.iium.edu.my/80015/1/FRGS17_031_0597IREP_IIUM.pdf; http://irep.iium.edu.my/80015/

Multiple highly manoeuvrable robots have become the focus of heated discussion lately, especially in applications involving disasterMultiple highly manoeuvrable robots have become the focus of heated discussion lately, especially in applications involving disaster rescue, military missions and underwater or extra‐terrestrial explorations. The surroundings concerned are harsh and hazardous terrains, and predictably the malfunction rate is high. Thus a highly robust autonomous robot that could manoeuvre not only on allterrain but also underwater is of prime importance. One of the most crucial aspects of the legged robot is to optimise the executing patterns of leg-joint angles which is known as gaits. Synthesising gaits by manual tuning control is a complex and time consuming task which becomes even more challenging when the vehicle operates underwater. However, the mechanism presented in the literature are focusing on individual mechanism either on-land or underwater gait configuration. In fact, both mechanisms are tackled separately whereas, in a typical application such as surveillance and disaster management, both mechanisms can be integrated and executed appropriately to fulfil bigger objectives in a better optimised relocate position, as well as in faster completion time. When operating on both land and underwater, any motion of the limbs could apply forces to the robot. Efficient gait identification must therefore be modelled to mitigate these unwanted forces while meeting the desired gait properties. This research proposed automatic gait synthesis system for a hybrid amphi-underwater robot. The system utilises an exponential momentum ant colony optimization which provides randomization and attractive parameter adaptation during the iteration process. The resulting will improve significantly not only manoeuvring time but also optimising total energy usage during manouvring stage on both land and on water.