A system for an anticipative front human following robot

Person-following by autonomous robots: A categorical overview

“…In particular, various aspects such as desired proximity behaviors, following angles (Shanee et al, 2016), turning and waiting behavior, etc., should be considered during trajectory planning. Additionally, application-specific choices, such as whether to stay behind or side-by-side while following, the desired speed, and relevant anticipative behaviors (Granata and Bidaud, 2012; Mi et al, 2016) are essential considerations. Another important feature is to maintain the expected behavior during explicit interactions (Hu et al, 2014; Islam et al, 2018c), e.g., being stationary when the human is communicating, and exhibiting correct acknowledgement responses.…”

“…Unlike in an underwater scenario, the person cannot come close to the robot and perform hand gestures in front of its camera. Consequently, the UAV might end up being too far away to detect various kinds of hand gestures (Bruce et al, 2016; Monajjemi et al, 2016). In such cases, it is often useful to use a reliable gesture (a static palm gesture, waving hands, etc.)…”

“…The visual challenges of detecting hand gestures from a distant unmanned aerial vehicle (Monajjemi et al, 2016): notice the minuscule appearance of people on the right compared with the left image where the UAV is much closer.…”

“…The choice of sensors is often determined by the operating environment, i.e., indoors or outdoors. For example, RGBD sensors are very effective in indoor environments; in addition to having a regular RGB camera, they are equipped with an infrared sensor to provide the associated depth information (Mi et al, 2016). Therefore, both the position and distance of the person can be measured with good accuracy.…”

A Cost-Effective Person-Following System for Assistive Unmanned Vehicles with Deep Learning at the Edge

“…Several works [15][16][17][18][19][20] use the first approach to approximate the distance information by triangulation methods applied on two or more RGB views of the same scene. However, the most used visual sensors for person detection are RGB-D cameras [21][22][23][24][25][26][27][28][29][30][31][32] that are able to get both RGB images and depth maps by exploiting infrared light. Several methods employ sensor fusion techniques to merge information from different kinds of sensing systems.…”