Low-cost scanning LIDAR architecture with a scalable frame rate for autonomous vehicles

Abstract

Autonomous vehicles need accurate 3D perception with a decent frame rate and high angular resolution to detect obstacles reliably and avoid collisions. We developed a low-cost scanning multichannel light detection and ranging sensor architecture allowing scalable frame rates by adjusting the number of laser and detector pairs. Scanning is achieved by a pair of micro–electro–mechanical system (MEMS) mirrors. A control pattern for the MEMS mirrors to maximize the frame rate is presented. A built prototype based on the proposed architecture achieves a frame rate of 11.5 Hz, a field of view of 70 ∘ × 30 ∘ , and an angular resolution of 0.4°. The distance resolution is 6 cm. Reliable single-shot detection for low-reflective objects up to 19 m indoors and 11 m under direct sunlight exposure is achieved. A performance assessment based on the presented measurement system for recently available vertical-cavity surface-emitting laser arrays with power densities up to 1 k W / m m 2 shows promising improvement potential.

Universal Behavior of the Image Resolution for Different Scanning Trajectories

“…Due to its great performance and low cost, MEMS LiDAR is quickly becoming an essential sensor in vehicle environment sensing systems. The LiDAR method can be used in many different areas, such as agriculture [14], autonomous vehicles [15], and drones [16].…”