Watch: AI robot patrols Dubai beach to monitor e-scooter violations

The robot runs autonomously but a safety officer is on hand to assist in case of emergency

by Angel Tesorero


KT Photos: Rahul Gajjar


A five-foot, 200kg robot, powered by AI (artificial intelligence) and armed with a wide camera and four motion sensors, is deployed by the Roads and Transport Authority (RTA) on Thursday to detect violations committed by users of bicycles and electric scooters along the beach of Jumeirah 3.


No fines will be issued yet to erring riders, RTA clarified, during the one-month trial period aimed at fine-tuning the function of the autonomous robot that is tasked with spotting violations, including failure of the rider to wear a helmet, having two or more riders on e-scooter, as well as parking and using e-scooters in unauthorised areas such as jogging tracks.


The robo-inspector, created by Chinese robotics and technical systems provider Terminus, runs autonomously, but a safety officer is on hand to assist in an emergency.


"For the next 30 days, the robot will sweep an area of about 600 metres along the jogging and cycling track of Jumeirah Beach 3, from 3pm to 11pm on weekdays and 5pm to 1am on weekends," said Hamad Alafeefi, director at RTA's Enterprise Command and Control Department (EC3).


Currently, the robot can identify violations with 85 per cent accuracy and surveillance reach of up to 2km. It can deliver data within 5 seconds to RTA and will share and analyse the information with Dubai Police.


Authorities can also calibrate the robot to make important public announcements such as call to prayer or to remind people to keep off the beach during inclement weather. The robot is designed to operate efficiently under various weather conditions.


The robot has intuitive sensors. It comes to a complete stop or sidesteps when it detects any object or individual within 1.5 metres in front of it — thus "ensuring the safety and security of road users in Dubai".




How will fines be given?

During the launch, the robot came to a halt as it encountered an e-scooter rider wearing no helmet. Since the roving robot is still in trial or proof of concept (POC) stage, there is no mechanism yet for how fines will be given.


Alafeefi underscored that the current objective is to detect and analyse violations in collaboration with the Dubai Police. "We need to ensure accuracy and lower error rate before going to the next level, so we can avoid complaints," he explained.


He added facial recognition could be an added feature that will be linked to Emirates ID to identify violators. The violations are punishable with fines of up to Dh300.




How to catch violators?

Currently, e-scooters are not required to be registered, although all riders are mandated to obtain a permit from the RTA.


Road safety experts suggested having a number plate or unique ID that will identify the e-scooter, aside from the rider securing an individual permit. "Registration will take away the anonymity of riders as both the e-scooter and rider can be identified in case of accidents (or violation)," Thomas Edelmann, founder and managing director of RoadSafetyUAE, earlier told Khaleej Times.


There is no updated information on the number of e-scooters across Dubai. However, based on the statistics released by RTA in November last year, the authority issued more than 63,500 e-scooter permits since it became a requirement in April 2022. Only those 16 years and above can ride e-scooters; permits are granted after completing training and awareness courses available on the RTA website. However, individuals who possess valid driving licenses are not required to take these courses, and those 15 and below can use manual scooters and bicycles.


How the robot works?

The RTA roving robot has an interactive function similar to those deployed during Expo 2020, shared Ehsan Hameed, sales director of Terminus Group in the UAE, the same company that deployed 150 robots at the Expo.


The robo-inspector has a dimension of 1,000mm X 800mm X 1,500mm (LxWxH). It has a 4K camera in front with night vision capability and motion sensors on all sides for 360-degree coverage.


The robot can run up to eight hours on a single charge depending on its speed (average is 7km/hour). Once it detects a low battery level, it will automatically retreat to a nearby charging station.




We use cookies to offer you a better browsing experience, analyze site traffic and personalize content. By using this site, you agree to our use of cookies. Visit our cookie policy to leamn more.