Visual support in target search for a simulated unmanned aerial vehicle

In steering Maritime Unmanned Aerial Vehicles (MUAV's) and controlling its camera movements, the operator's task is complicated by a limited quantity and quality of the available perceptual information. For example, the outside image is presented with low update rate and a narrow field of view due to the limited bandwidth of the data-link between platform and operator. The reduced amount of visual information may hamper visual search, target track in this connection the TNO Human Factors Research Institute has developed a new principle of visual support of which some of its potential benefits were demonstrated in a previous experiment. (Van Erp, Korteling & Kappé, 1995). This principle involves an earth-fixed Computer Generated Grid (CGG) consisting of parallel and perpendicular lines, depicted with a high update frequency over the (slower updated) camera image. The current experiment focused on the benefits of the CGG in improving target search performance, and com-pared its performance with other methods of operator support. In a search task, subjects had to locate five target ships on a, further empty, sea as fast as possible. Information on the location of the MUAV and the target ships was presented on a simulated radar image.
The subject could be assisted by two different types of visual support: an earth-fixed CGG at sea level, or two head-up linear quantitative indicators adjacent to the camera image, indicating camera pitch and compass heading. As baseline, all conditions in-cluded a standard method of operator support by means of a pictorial combined heading and pitch indicator (CHPI).
The information provided by the CGG is fundamentally different from the information provided by the indicators. Gibson (1950) assumes that the information provided by the CGG may be picked-up directly by the visual system, without demanding substantial visual attention. The more traditional methods of operator support all require the opera-tor to use some kind of cognitive strategy to infer the MUAV attitude from the pre-sented abstract information.
In order to gain more insight in the potential beneficial effects of the two types of operator support, the MUAV could have a fixed position in the air, or could fly a circular flight path. It was expected that a circling MUAV would complicate perception and discrimination of camera and MUAV motions. Also, the MUAV could be positioned central or eccentric in the radar dispradar dispric in the radar display. Only when the position of the platform is in the centre of the radar image the quantitative indicators and the CHPI provide straightforward information. In an eccentric position, subjects have to calculate in order to determine the correct heading of the camera by using the compass scale.
The results clearly substantiate the effectiveness of the CGG in improving the operators search performance: search time and total camera-heading and pitch movement were significantly reduced when the CGG was presented. Supporting the operator by means of quantitative indicators, depicting camera heading and pitch, did not reach significance on any of the dependent variables.