SBIR-STTR Award

K2T, Inc. proposes to design a walking machine for shallow underwater mobile operations. Development of an underwater walking robot is important due to the need to explore ocean bottoms for oceanography, bottom tomography, performing pollution assessments, and ship inspection. Shallow water in particular is a difficult portion of the ocean to explore due to the heavy currents and surges common in coastal waters. The objective of the Phase I research is to design a walking machine which can operate in the presence of subtantial fluid perturbations. K2T will draw upon its experience as the mechanical designers of the Antarctic volcano exploration robot, Dante, to create a design which overcomes much of the difficulties associated with fluid perturbations by emphasizing stability, a conservative gait, and a straightforward vehicle controller. The results of the Phase I effort will be a feasible design including system configuration, formulization of walker controller, performance specifications for legs, kinematic simulations, force stability study, gait study, and feasibility assessment based on biological principles and comparisons to predecessor walking machines. This design will provide a foundation for the Phase II construction of a prototype walking vehicle.

Keywords:
legged robot walking machine underwater robot

K2T, Inc. proposes to extend concept work carried out under Phase I funding to produce a prototype version of a mine removal walker. The proposed walking machine will be capable of autonomous operation in the surf zone of expected amphibious coastal landing areas. The MRW will contain an explosive charge, and will search for mines using internal sensing. Upon mine detection, an MRW will walk towards the mine and maintain pollution in the vicinity of the mine until instructed to detonate. The design provides for a robust, low cost solution which uses a strategy in which mis-steps and roll-overs can be easily handled. A control structure has been designed which uses distributed processors to implement individual behaviors. A rudimentary set of body behaviors is used as a foundation to develop more complex behaviors such as obstacle handling and mine seeking. K2T has partnered with Case Western Reserve University who has experience with the design and neural-based biologically-inspired walking robots. A prototype walking robot will be demonstrated at the end of the Phase II effort. The prototype will serve as a demonstration and research platform for further Phase III development and fund acquisition.