Underwater archaeology experiment

At present, SPC-II can elementarily fulfill the need of underwater archaeology and it also has some advantages in environment disturbance and horizontal maneuverability. But the ascending and submergence of SPC-II are implemented by changing the pectoral fin's incidence angle, so it can dive only in the process of swimming (submergence velocity is the minimum straightforward velocity of submergence). In addition, it can not swim back off and hasn't the underwater position system. All of which bring more difficulty to underwater pointing task. Moreover, SPC-II is only a robot prototype with simple equipments; it can only help the archaeological team to complete the second phase task. Donggu gulf site of shipwrecks in Dongshan city, Fujian Province is in a gulf, and the offing is much calm, but the water quality is much turbid (especially when is ebbing). The position of site has been known that is in less than 5 meters. The benthal hypsography likes gently brae, and the diffused china, iron weapon and ships were enchased or buried in silt. The purpose of our attempt is to let the robotfish dive to the site and capture and feedback the valuable underwater image information by the vidicon. The implementing methods are described in detail as follows.

1) Experiment scheme of robotfish aided underwater archaeology

The communication chain between the SPC-II and the console was properly modulated. As a result, the communication with cable and the wireless communication are both adopted (buoy and radio-communication as shows in figure. 5). This method not only saves the expensive underwater communication sonar, but also avoids the encumbrance result of too long cable.

antenna console antenna console

Fig. 5. Sketch map of system communications

The method of hover submergence to explore the object was developed, which would resolve the problem that SPC-II could not hang when swimming and could not ascend and dive vertically. Because the turning radius of robotfish is small and the spiral velocity is slow, and the robotfish can keep swimming in stable depth, it can screen the object distinctly from any directions. The process can be divided into two steps: 1) arriving aim area; 2) scanning and searching unidentified objects. It can be decomposed into several sequences of robot actions or control operations. First the robotfish swam on the sea level at high speed to reach the above of the aim area with GPS navigation; then it spirally dived to appointed depth in short turning radius (Figure 6 (a)). And now the robot had reached the aim area. Two measures were implemented for scanning and searching unidentified objects. The first is that the robotfish hovered in stable depth at diving velocity and gradually augmented the hover radius along the involutes (Figure 6 (b)). Moreover the second is that it swam out and home in stable depth at diving velocity and gradually made lateral move (Figure 6 (c)). The robofish started the video module to capture and the console started to monitor synchronously and recorded the video information during the process mentioned above. Once discovering the questionable objects, the robot hovered at the place and modulated the submergence depth in order to receive more clearly and abundant image information. But in practice, the sea-bottom gradient is changing and SPC-II does not have the ability of landform tracking, so the searching effect in stable depth (3 meters) was not perfect, and it usually appeared the phenomenon that the level of sea-bottom had overstepped the camera's eyeshot. Consequently, the method, that SPC-II swam in the ordered height little higher than sea bottom and explored the sea-bottom all the time, was proposed. It had been proved to be much fit for the even landform.

(a)hovering (b) hovering scan (c) traversing scan

dive

Fig. 6. Methods of robofish aided archaeology 2) Experimental result

The practice has proved that the strategy mentioned above is feasible. The robofish can dive to target area to search and can cover it on the whole, at the same time, the images of target are returned . But the efficiency is low and the eyeshot of single image is too narrow, which could not form the panorama vision yet. Comparison of the two measures for searching questionable objects has shown that the efficiency of the first is higher than the second in theory, but its control is complex, and the robofish is hard to swim according to the scheduled track, which could result in omitting repeating scanning. But the operating of the second is easier compared to the first. Table 2 gives the result of several methods of searching for an square area of about 200 square meters.

Search Method

Time

Number of

Consumed

dubious site

Was this article helpful?

0 0
Learn Photoshop Now

Learn Photoshop Now

This first volume will guide you through the basics of Photoshop. Well start at the beginning and slowly be working our way through to the more advanced stuff but dont worry its all aimed at the total newbie.

Get My Free Ebook


Post a comment