**4. Methodology and Research**

The research presented in this paper aimed at providing empirical verification of the automotive anti-collision radar for target detection in a water environment. For fulfilling this aim a set of research was designed with the use of empirical measurements and statistical data evaluation. The radar device was mounted on an ASV and it was used in real time in pre-planned scenarios (stationary and including movement of the platform). The data was recorded and in a later processing stage, statistically analyzed. This section provides a description of the research concept, scenarios, research equipment, and statistical data evaluation, while the results are given in the next section.

#### *4.1. Research Concept and Scenarios*

To collect robust measurements, useful for evaluation, the study first identified the detection possibilities and range for the objects that were typically in inland waters and then determined the empirical field of detection, which was verified with a declarative beam pattern.

The data was collected in two scenarios, as stationary research and on a moving platform. In both scenarios, the radar was mounted on the HydroDron ASV, which is described in more detail in the next section. The research was conducted on Klodno Lake in northern Poland. In the first part, the HydroDron was moored at the end of the wooden jetty and the targets were moved to provide observational data for various targets and to characterize their detection parameters (see Figure 4).

**Figure 4.** Configuration for stationary research. Targets were moved along the "moving lane" in front of the stationary HydroDron.

The targets used in the first scenario are presented in Figure 5, including an airtoy (dragon), lifebuoy, small boat fender, swans, and a radar reflector. These targets were selected as typical objects that could act as obstacles on the water surface. The initial research undertaken for the moving ASV has shown that the detection ranges were small for most of these targets. Therefore, better characterizing their detection in the stationary scenario was deemed important. Additionally, detection of the floating radar reflector was tested to determine if the shape of the target had an influence on detection. It has to

be pointed out that even such small objects can damage or impair valuable devices carried on board the ASV or the vehicle itself.

The second scenario was performed with a moving ASV and one target, a moving inflatable boat (Figure 6), as an example of a typical collision target on inland waters.

The goal of the second scenario was to analyze and characterize the detection while the target was moving in a typical way. The scenario was divided into three stages, corresponding to IMO Collision Regulations—head-on, crossing, and overtaking. During crossing and overtaking, different ranges were tested to obtain complex information regarding the angle of view. In general, more than 40,000 single radar measurements were collected and analyzed, representing 5 head-on, 11 crossing, and 10 overtaking situations.

**Figure 5.** Objects used in the first scenario as targets, (**a**) airtoy, (**b**) lifebuoy, (**c**) fender, (**d**) swans, (**e**) radar reflector.

**Figure 6.** Inflatable boat used in the second scenario as the target.
