An AR-HUD (Augmented-Reality Head-Up Display) overlays digital information onto the real environment. Many AR-HUD studies focus on providing virtual image depth (VID): some approaches present multiple depth layers (often at fixed depths), while others enable dynamic VID with a limited number of layers. As a result, achieving hologram-like, freely controllable multi-/variable-depth imagery remains challenging.

Meanwhile, SLM-based holographic displays, which can offer multi-/variable-depth imaging, typically have a narrow viewing window; when the viewer’s head moves and the eyes leave this window, the holographic image disappears. Toward SLM-based AR-HUD, we developed a compact 3-DOF manipulator that steers the projection based on pupil positions, keeping the holographic image visible over a wider viewing range. Because light must be delivered to both pupils independently, we designed two closely packed manipulators that operate independently while maintaining a compact.