Thought leadership article by Bernard Dion, Fellow and Chief Technologist for Systems and Embedded Software at Ansys.
–//
Across the aviation industry, advanced air mobility (AAM) is becoming a major focus for original equipment manufacturers (OEMs) and suppliers alike.
Companies are developing aircraft for people and cargo that will aid underserved local, regional, urban, and rural environments. And to make these innovations as efficient and effective as possible, autonomy has become a key focus.
This next generation of aircraft will be smaller, and will fly at lower altitudes. Whilst some are being based on existing technologies, such as helicopters, a competitive market for electric vertical take- off and landing (eVTOL) aircraft has emerged. Research around AAM in the UK estimates that up to 11.4 million journeys will be substituted by eVTOLs by 2040, resulting in up to £2.1bn in annual socioeconomic benefit.
The widespread adoption of autonomous AAM has the potential to revolutionise transport, and the industry, as we know it. Consumer demand already exists, with 57 per cent of people across the globe open to travelling in eVTOLs.
Yet, there is still some way to go – getting an aircraft in the air safely is the highest priority, and regulation authorities have strict standards that companies must meet. As autonomous technologies develop, upholding safety standards becomes even more crucial. To adhere to regulations and decrease time to market, engineers can establish a digital thread, powered by simulation.
Addressing challenges around autonomy in AAM
Autonomy is based on three functions. The first is perception based on sensors like cameras, lidars or radars. The second is decision-making to avoid a collision by detecting an obstacle, for example. The third is actuation, which uses flight and engine controls to perform the correct manoeuvre without human intervention.
Developing these autonomous functions, and ensuring their safety, is an incredibly complex and intricate process. Engineers must perform a safety analysis to analyse the operational design domain (ODD), and outline the challenges faced during development.
A multitude of factors, such as the weather and electronics failure conditions, must be taken into consideration to ensure that AAM aircraft can operate effectively. For example, extreme weather conditions like fog can impede sensors and risk poor perception and decision-making. Electronic failures could also occur, which can impact the actuation function and result in the wrong manoeuvres.
This analysis is the most critical and difficult factor in the development process, so engineers can turn to simulation to overcome potential safety challenges. Simulation can be used to digitally replicate and test any component of an aircraft such as the powertrain or aerodynamics. This provides a clear, 360-degree view of how the aircraft will operate in different environments.
Additionally, engineers can easily create an endless number of potential scenarios that a vehicle may
encounter, allowing them to analyse how it would react, and then optimising accordingly to guarantee safety. By digitally replicating the testing process, engineers can easily pinpoint and rectify potential errors to create a thorough safety analysis to meet regulation authorities’ standards, and decrease the time to market.
Advancing safety in autonomy with digital engineering
Simulation is pivotal for developing and safety testing autonomy in AAM, but developing autonomous functions requires interconnectedness. Engineers must therefore incorporate simulation into an overall digital environment to achieve breakthroughs in autonomy and keep up with emerging safety regulations. With digital engineering, engineers can connect simulation to the processes, tools, and data across product development, and can ensure safety in complex products.
This way, they can easily collaborate to virtually design and test a product. Creating a digital environment establishes a digital thread, which enables traceability and effective collaboration between the AAM operator and the aircraft developer.
This is vital to maximising passenger safety and meeting regulations, as digital engineering impacts more than just the development period. A digital thread allows engineers to monitor and update the product to ensure it operates correctly in the event of an error. As AAM evolves, teams can also easily ensure that the product continually meets developing safety standards.
The role of artificial intelligence
As artificial intelligence and machine learning (AI and ML) become increasingly integrated into
autonomy functions, digital engineering can also help train the technology to maximise safety.
AI and ML are being leveraged to develop the perception function in place of complex image processing to perceive the aircraft’s environment. They’re also being used in place of complex control laws to improve decision-making in uncertain environments.
This added layer of intelligence will accelerate autonomy in AAM, as long as it’s developed in an interconnected digital environment. With digital engineering, simulation can be leveraged to perform supervised learning by providing images and solutions to train AI and ML so that they can develop the perception function.
Simulation can also train AI and ML in the decision-making function through reinforcement learning, a trial-and-error process. Engineers can then see how this process will impact the overall safety of the aircraft, bringing them one step closer to the widespread adoption of AAM.
The future of AAM
The aviation industry is on the precipice of transformation. Companies are working together to build the future of flight – but safety is paramount to the development of these AAM aircrafts. By leveraging simulation and establishing a digital thread, engineers can ensure that these innovative new technologies meet evolving standards and regulations, so that the future of AAM is as safe as possible.
–//
You can listen to more of Bernard’s insights by listening to our latest podcast, which was released last week and is available on whichever podcast platform you use.
