MIT Ocean Engineering - RoboTuna

Link: MIT Ocean Engineering - RoboTuna.

2008/10/20 updated with a comment pointing to a weblog entry with additional information about RoboTuna and complementary work by Prof. Joseph Ayers

This website contains a wealth of information about the MIT program to emulate the swimming dynamics of the bluefin tuna.  In addition to pictures of RoboTuna II, the website includes information about the original Charlie I, developed by David Barrett for his doctoral thesis.  His goal was supporting long range/long duration Autonomous Undersea Vehicles by "produc[ing] a dynamic body motion that can realistically recreate the type of flow field that exist about and behind a swimming biological tuna."

The Origins of the MIT RoboTuna describes the challenges of emulating the complex dynamics of a biological tuna.  Although only a low-order copy, an analytical model of tuna movement is currently not possible due to the complexity of interaction between the undulating body and the surrounding fluid. 

The Design of the MIT RoboTuna describes the process of abstracting the fundamental principles of the system and then develop a physical implement using available technology.  The model design and structural components replicates the major features of a tuna.  The model can be tuned to better understand the impact on drag and thrust. 

Past and Future Work on the MIT RoboTuna describes the use of a Genetic Algorithm to efficiently search for optimal parameters.  The result supports Gray's Paradox: drag on the swimming RoboTuna is less than on the motionless model.  The mechanism is not fully understood but does not require any unique power source.  So far, the RoboTuna has reduced its 'base' drag by 50%, while Gray's paradox suggests a seven-fold reduction may be possible. 

Future work includes:

• flow visualisation and modelling
• incorporating a Tuned-Harmonic-Drive
• supporting free swimming

Comments

[NH]

The weblog entry 'Aquatic robotics' (http://kr.blog.yahoo.com/ghleeblog/3347) provides additional information on the efficiency of Robotuna as compared to propeller-driven vehicles and the computer used to control it (a "flexstack" computer the size of a coffee cup).

It also mentions the work of Prof. Joseph Ayers who is developing RoboLamprey and RoboLobster. Ayers is a leader in developing simplified but effective real-time control systems that require modest amounts of computing power.