Published on

E4 the future of indoor spaces protection

A.I.MERGENCE is a start-up specializing in Artificial Intelligence and Autonomous Robotics, it designs E4 a mobile safety robot

Description

There are still very few robotics companies in France today. In addition, we have created the first robot dedicated to indoor spaces protection. We have chosen to design a robot specialized in the field of security, aiming to best match the requirements of its users. Our will is to create a quality robot, with the best embedded technologies and the best possible features in the field of indoor monitoring. In addition, its design (size: 35cm span) is best suited for small spaces. Its small size and agility allow it to navigate in complex environments. It knows how to be discreet when it needs to, but also to be dissuasive in the event of an intruder trying to break into your home. Thanks to its mapping system, it knows its location and therefore the area it needs to monitor. It is also equipped with a facial, morphological and voice recognition system, which allows it to recognize owners, intruders, animals, etc. You will no longer need to worry when you are away, E4 will notify you in case of an incident. In conclusion E4 can monitor your house, detect and deter an intruder, detect fire, water leak or another anomaly such as a broken window for example. It moves in the spaces and is totally autonomous. This means that no settings are necessary, no sensors or cameras are to be installed in your house or room. We benefit from real transparency in use since we no longer need to think about turning it on or off to make it work. Even at night no worries to have it watch over you. Problems with existing alarm systems today: - The installation of such a system is complex and requires the intervention of a qualified person. - A multitude of sensors and cameras must be installed in the areas to be monitored (intrusive) - The current cameras do not have a 360° vision system and are therefore constrained to a limited viewing angle. - The current systems are not autonomous and require an action from a human (starting of the system, activation thanks to a button, a door code etc.). - Few systems today are equipped with facial recognition and do not allow identification of persons authorized or not to enter a space - Some alarm systems use an online storage system (cloud) which increases the risk of hacking from outside. A hacker can therefore observe through the cameras what is happening inside. - When a person returns home, they must automatically deactivate their alarm system. This no longer allows him to ensure the security of his home in the event of intrusion, whereas in 30% of cases during a burglary at least one person is present. Solutions provided by our E4 robot: - No settings are necessary, once the robot is activated in the house it will be able to self-manage, recharge itself, locate itself in the space and learn from its environment (100% autonomous, ease of use) - More than facial recognition, our robot is equipped with multimodal person recognition (facial recognition, voice recognition, morphological recognition, etc.) This makes it possible to obtain better results in terms of identifying a person at a lower cost. - No untimely triggering in case of detection of a pet. - No untimely triggering if an authorized person enters the secure space (maid, staff, family - friends etc.) - It also detects anomalies such as an open or broken window, a fire, a water leak, etc. - In addition to its preventive action (sending notification in the event of intrusion), it can be dissuasive by triggering blinding flashes of light and by triggering its own alarm system of 95 decibels - The data is stored directly in the robot, no cloud computing system and therefore no hacking by a third party possible. - The robot works day and night, so it ensures the safety of its user 24 hours a day.

References

E4 is programmed to detect intruders, fires, water leaks, or anomalies. It is able to locate them thanks to different components: For person detection: - Microphones o Location of suspicious noises (the robot moves directly on site to check the area) - Top Fisheye Camera (360°) - Front HD Camera  We also use algorithms to identify people to know if they are intruders or not. Action performed in case of intruder detection: - Blinding flashes - Buzzer alarm (95db) For fire detection: - Microphones (fire alarm identification) o Sending notifications in the event of a fire alarm being triggered - HD camera, Fisheye camera (360°) o Flame recognition algorithms - Temperature sensors For water leak detection: - Humidity sensors For the detection of other anomalies: - Microphones - HD Camera - Fisheye camera (360°)

Attachments

Evaluations