Automating the Hyperloop

The transportation sector is an industry heavily reliant on operational staff. Cars require a driver to be on the road, airplanes cannot function without two pilots and a train uses a driver for acceleration and braking. However, in the last decades operational tasks have slowly become automated. The shift towards autonomy is not only relevant for current modes of transportation. This article will focus on a certain type of automation: Artificial Intelligence (AI) in the Hyperloop system. The current applications of AI in transportation will be discussed, an overview will be given on the possible uses in the Hyperloop and the requirements for AI will be examined. 

Artificial intelligence is not a single technology but rather a collection of different approaches, methods and technologies all showing intelligent behaviour to various degrees (EPRS, 2019). AI has heavily influenced many industries and the transport industry is no exception. Cars are becoming more autonomous every year and the self-driving vehicle has become a realistic concept. Many metro systems around the world no longer require a driver to operate. Trains use AI applications for operations and logistics. This shift from human interaction to machine-based decision-making is therefore important for designing the Hyperloop system. 

Artificial Intelligence can be used in various forms. The applications used for designing, operating and maintaining the Hyperloop system will be discussed briefly: 

  • Designing the Hyperloop network and stations: Designing a network is complex when many different station locations are considered.  By using the demand per area and infrastructure costs, AI can iteratively design a network which uses cost-effective links to transport as many passengers as possible. Moreover, the link and network capacity can be used to design Hyperloop stations with the appropriate size to handle different passenger flows. 
  • Incident detection: Hyperloop safety is an important topic and the prevention of accidents is one of the main priorities. Monitoring the pods, the infrastructure and the facilities allows for the discovery of the causes of accidents. Artificial intelligence can analyse the data gathered and detect anomalies when the actual data deviates from the predicted data. By alerting authorities of these anomalies, future incidents can be prevented. Besides using data analysis for detecting technical failures, it could be used to improve station security. Passenger information and security footage can be used to analyze crowds and alert security officials on potential threats.
  • Automated operation: The Hyperloop is designed to function autonomously. This includes not only automatic acceleration and braking but also reacting to problems and hazards during the trip. The implementation of AI allows the automated operation system to make quick decisions during emergency situations therefore improving passenger safety. 

The value of AI is clear for the Hyperloop system. However, certain requirements need to be met before the use of artificial intelligence in the Hyperloop can become a reality. The first requirement is that AI uses a lot of computational power. This computational power needs to be supplied before the automated tasks can be completed. The purchase costs of the systems and the operational energy use is high, so a careful trade-off needs to be made between costs and value of the AI service. Moreover, there is the issue of data. AI is capable of analysing data, gathering useful insights and using those insights to decide on a course of action. However, for a new system such as the Hyperloop, this data does not yet exist. Therefore, it is important that, from the first tests until operational use, as much data as possible is collected about  Hyperloop infrastructure and pods. 

European Parliamentary Research Service. (2019). Artificial Intelligence in Transport. 

Leave a Reply