Compact 3D Time-of-Flight Sensor with Real-Time Multi-Path Separation

The acquisition of 3D data using indirect 3D Time-of-Flight (3D ToF) sensors is a well-established technology. How-ever, Multi-Path Interference (MPI) is a key issue and causes significant errors in the distance measurements. This paper presents a novel compact 3D ToF sensor with real-time multi-path separation. The sensor combines multiple measurements at different modulation frequencies and employs Particle Swarm Optimization (PSO) in combination with Sequential Quadratic Programming (SQP) to separate the different paths. The sensor is used to demonstrate the separation of two different paths in each pixel of the entire image in real-time. Evaluation in various laboratory and real-world scenarios reveals a significant improvement in distance accuracy compared to a standard 3D ToF sensor. Furthermore, the proposed algorithm exhibits processing speeds that are orders of magnitude faster than previously reported separation algorithms with comparable accuracy.