Ad-hoc collision avoidance system for Industrial IoT

Safety and security in the workplace are extremely important. In the industrial sector, with heavy machinery, an
accident can have serious if not life-threatening consequences. In Industry 4.0 with the inclusion of paradigms
such as the Industrial IoT, new technologies can help not only to improve the production process of a company but can also assist by improving the safety and security of industrial workers. Security is sometimes considered an add-on or something to be applied in the future, which could lead to very expensive or complex solutions. In this work, we deal with scenarios where heavy machinery and workers are in a common area. In these scenarios with heavy machinery, accidents can be caused by different factors. To aid in the avoidance of such incidents, we create a system that detects the presence of nearby workers by using Bluetooth devices. Using the signal emitted by these devices, the driver of the heavy machinery is warned through visual and haptic signals. The system is designed to capture the attention of drivers and their peripheral vision with very simple signals, easing the learning curve of the system. The contribution of this work also provides an implementation the system using Bluetooth for the emitters and the detectors, which are connected to a Raspberry Pi that manages the ADAS, evaluating the information about surrounding Bluetooth emitters, and triggering the different feedback mechanisms as required. Accuracy is a limitation of Bluetooth-based location systems. Nonetheless, the proposed system allows us to detect nearby workers relative to the heavy machinery, as opposed to pinpoint their location in a specific area. Furthermore, the system is very affordable and minimally intrusive compared with the use of cameras that require that the driver’s attention is drawn away from the driving process. In this work, we design and provide the proof of concept of the complete collision avoidance system.