Join or Sign in

Register for your free membership or if you are already a member,
sign in using your preferred method below.

To check your latest product inquiries, manage newsletter preference, update personal / company profile, or download member-exclusive reports, log in to your account now!
Login asmag.comMember Registration

Taiwan had a terrible train accident. Here’s how AIoT can help prevent another one

Taiwan had a terrible train accident. Here’s how AIoT can help prevent another one
Roughly a month ago in early April, Taiwan had one of its worst train accidents in history. This note examines how advanced technology such as AI and IoT can help improve rail safety and prevent accidents like this from happening again.
Roughly a month ago in early April, Taiwan had one of its worst train accidents in history, with 50 deaths and over 200 injuries reported. This note examines how advanced technology such as AI and IoT can help improve rail safety and prevent accidents like this from happening again.
The accident occurred in the morning of April 2. A train traveling at over 100 kilometers per hour crashed into a crane truck that fell on the track immediately before a tunnel. The train then derailed and piled up in the tunnel. The truck fall was due to human error.
Indeed, it’s extremely difficult for any driver to respond in such a short time. “Apparently, the video from the train shows that the driver had about 6.9 seconds to respond to a blocked track condition about 250 meters away. Human reaction time (let alone other factors such as train braking performance) at the speed that the train was traveling was simply not going to be good enough to avoid the accident,” said Syed Zaeem Hosain, CTO at Aeris Communications.

Investigation findings

The accident was investigated afterwards. One of the primary investigators was the Taiwan Transportation Safety Board, which deduced that the crash occurred less than two minutes after the truck fell onto the track, Taiwan’s Central News Agency reported.
The question, then, becomes if technology is available to alert trains that there is large-size debris or obstacles ahead. The answer is yes. The technology is in the form of Internet of Things and AI, which entail data generation and processing to provide situational awareness to the user.
“Yes, the technology to solve such problem is available. IoT, video surveillance system and other infrared sensors are there to detect any intrusion of any foreign object (may be any other moving vehicle, bolder, big tree branches, wild animals such as elephant) on the track,” said Sanjay Kumar, Chairman of Railway Recruitment Board at Indian Railways. “Information and communication technologies (ICTs) play a significant role in disaster prevention, mitigation response and recovery. Timely, predictable and effective information is required in rescue operations and decision-making processes.”

How AIoT can help

Detection of foreign objects relies on different types of sensors under the IoT scheme. Obviously, IP cameras along the track are important sensors. Radar sensors can also come in handy, detecting abnormalities during nighttime or in inclement weather conditions.
Remote sensing can be helpful as well. Irregularities can be detected by communication satellites and/or drones. The latter, equipped with various types of sensors, can also play a role in predictive maintenance.
“Remote sensing as a tool can very effectively contribute towards identification of hazardous areas, monitor the planet for its changes on a real time basis and give early warning to many impending disasters. Communication satellites have become vital for providing emergency communication and timely relief measures. Integration of space technology inputs into disaster monitoring and mitigation mechanisms is critical for hazard reduction,” Kumar said.
More and more, detection is done by LiDAR sensors, which measure distance by shooting the target with infrared laser and measuring the time of flight (TOF) using a wavelength-sensitive sensor.
“LiDAR is emerging as a preferred approach, especially those systems with long-range detection or classification capabilities needed to alert a high speed or freight train in time to prevent a trackside incident,” said Keith W. Dierkx, rail industry executive and advisor to LiDAR provider AEye.
Akram Benmbarek, VP of BD and Strategic Initiatives at AEye, added: “Long range LiDAR is also the most straightforward solution to prevent such accidents. The ability to detect an object on the track at 1,000 meters can lead to more preventive options, even if it doesn't get the train to make a full stop. While the train is slowing down, it can send warning signals to move the object, or have a preemptive response.”

[LiDAR vs. camera-based people counting: which is better?]
In the event of a train accident, reporting this to first responders to minimize casualty is critical. In this regard, automated crash notification systems working in conjunction with different sensors, including acceleration sensors, can be an effective tool.
“Such automated crash notification (ACN) systems in the connected car industry are already in use to provide rapid data to first-responders. These ACN systems could be adapted for other transportation in motion – for example, passenger railroad cars,” Hosain said. “The main point is that getting data to first-responders rapidly is vital – the faster that an effective first-responder response is provided, the more lives that can be saved.”
Indeed sensors can produce data. Yet AI is the brain that processes this data, making sure that obstacles and abnormalities are detected in the most timely and accurate manner.
“The use of AI in public transport appears to be one of the critical solutions that efficiently unlocks the value of data to improve the quality and efficiency for the public transport sector, especially in railways,” Kumar said. “AI-powered video analytics is considered to play a significant role in improving capabilities of railways’ predictive maintenance regimes (or condition-based maintenance) and security management.”
“AI has a preventive benefit if used with near-miss incident data, in a sense that it can predict when these accidents will happen in a specific area, which leads to extra precaution in that prone zone. LiDAR can play a major role in collecting data about near miss incidents,” Benmbarek said.

Some afterthoughts

Needless to say, deploying AIoT has its share of challenges. It requires an upgrade, if not overhaul, of the entire rail infrastructure, and the cost of deployment can be quite prohibitive. Yet in the end, rail operators will come to the realization investing in this technology will ultimately be conducive to improving rail safety. Hopefully, advanced technology like AIoT will see more deployment in rail transportation and help prevent another train tragedy from happening.

Product Adopted:
Subscribe to Newsletter
Stay updated with the latest trends and technologies in physical security

Share to: