In certain multi-sensor setups where there are overlapping fields of view, laser interference can occur when the camera from one sensor exposes and detects the laser from another sensor––resulting in what is known as “crosstalk” between sensors. This blog will discuss how Gocator® solves this challenge through a technique called exposure multiplexing.
When To Multiplex
A common requirement for multi-sensor systems is to precisely control the timing of each sensor in order to avoid laser interference in areas where the measurement zones overlap.
The solution to this problem is to identify sensors that are independent of each other (that is, without crosstalk issues) and logically group them into banks. Measurements in the system are then orchestrated so that each bank takes turns exposing the object, that is, the banks are multiplexed.
Exposure Multiplexing in Gocator®
Gocator® offers a built-in Exposure Multiplexing setting, which is easily enabled/disabled in the Gocator® user interface.
When enabled, Exposure Multiplexing creates a time delay for laser exposures and ensures that interfering lasers are not strobed at the same time. As a result, sensor crosstalk is eliminated and only the right data is acquired from the scan target. The Exposure Delay parameter in the sensor configuration is the key mechanism in achieving the various multiplexing behaviours supported in Gocator®.
The overall system speed required to solve the application at hand will determine the length of the time period for one data acquisition cycle, the so-called frame period. Generally, the frame period is set to the same value across all sensors in the network. This frame period is then split up into shorter time slots, one for each bank of sensors to operate within such that their exposure periods do not overlap.
The characteristics of the target surface (e.g., log, rubber tire, pavement, metal housing) typically determine the minimum exposure time needed for a reliable measurement (e.g., longer exposure for lower contrast objects). All sensors in the system must run with the same exposure time period and exposure mode (Single, Dynamic, or Multiple).
In order to control exposure down to 1 microsecond (µs) in a multi-sensor system, LMI offers a Master hub that looks just like an Ethernet switch but handles the distribution of power, laser safety interlock, and high speed synchronization. Hubs offer 8 to 24 sensor connections and can be daisy-chained to support larger multi-sensor layups.
The synchronization data transmitted by the Master to all sensor ports encodes a global time stamp (accurate to < 1µs), encoder stamp, and status of discrete inputs (eg., photocells connected to the input I/O of the Master). This synchronization data is received and decoded by each sensor to drive global exposure control and triggering logic. Each scan taken by a sensor is stamped with time, encoder, and input I/O status, which is later used by software to build a single 3D model from multiple sensor data streams.
The basic principle to avoid crosstalk in a multi-sensor network is to use a multiplexing strategy. Multi-sensor networking with multiplexing is supported by every Gocator® 3D smart sensor, allowing for high-precision 3D scanning and measurement in challenging applications where there are overlapping fields of view.
For more information on this subject, download the full Application Guide.