Characterization and System Identification of an Unmanned Amphibious Tracked Vehicle

Experimental testing of an unmanned amphibious tracked vehicle on dry sand, in the surfzone and in water, has been performed to explore its maneuvering and performance characteristics for the planned future development of an automatic control system. The 2.69-m-long, 295-kg concept vehicle utilizes a small waterplane area twin hull (SWATH) configuration with integrated crawler tracks and twin propellers for propulsion. Maximum traversable incline tests on dry sand reveal that the vehicle has a drawbar pull (DP) of about 1000 N (about one third of its weight) and can operate on slopes of 15 briefly and 10 for extended periods of time. Maneuvering tests were performed on flat, dry sand, as well as into and out of a surfzone at a beach site with inclines ranging from about 3 to 6. In each of the tests, the track forces, speed over ground, and both linear and rotational accelerations were measured. Also recorded were environmental conditions, such as wind speed, significant waveheight/period, and ground slope. The tests reveal that the vehicle has a maximum straight-line flat ground speed of about 2 m/s, a minimum flat ground turning radius of 2.4 m, an in-water minimum turning radius of 1.9 m, and a maximum straight-line waterborne speed of 1.2 m/s. It is also shown that system identification applied to the data recorded from the flat ground and waterborne maneuvering tests can be used to find a linear, parametric state-space model for the vehicle that adequately reproduces its motion. Land-to-sea transition tests in two different seas of one-third significant waveheights of 8 and 28 cm (both measured at a distance of 6 m from the mean waterline on the beach) show that the vehicle exhibits substantial track slip as it traverses into the surfzone and its weight becomes increasingly supported by the displacement of its hulls. The tractive force model developed by Wong (Theory of Ground Vehicles, New York, NY, USA: Wiley, 2001) is modified to account for the reduction in track-supported vehicle weight as the vehicle becomes waterborne. It is shown that this adapted model captures the main physical features of the measured track forces. The waveheights and periods recorded during surfzone transition tests are used to examine the seakeeping properties of the vehicle. It is found that the vehicle's natural frequency of roll is near the dominant wave frequencies measured. A Froude-Krylov strip theory simulation shows that the wave forces acting on the vehicle in 28-cm waves may be slightly larger than the force measured on each track when the DUKW-Ling is crossing the transition zone between the beach and surfzone and should not be ignored in modeling and simulation studies.