What are the challenges in the way of autonomous driving

The technology necessary to enable self-driving vehicles is evolving at an extremely high speed, but there are still multiple challenges in the way. 

A few days ago, when Spain’s technology conglomerate GMV announced that it had obtained a contract from BMW for developing a precise GNSS positioning system with integrity for the new generation of autonomous vehicles of the German carmaker, the long-standing discussion over the practicality and viability of self-driving vehicles resurfaced. Apart from technology, there are many other impediments in the way of effective deployment of autonomous mobility, such as challenges related to road safety legal framework, standardization and certification frame, and road and telecommunications infrastructure. According to GMV, a privately-owned technology business group founded in 1984, in addition to high accuracy required for the applications and functionalities involved in autonomous driving, it’s the safety incorporated to the concept that is crucial.

The technology which is necessary to enable autonomous cars is evolving at an extremely high speed, and so, making predictions is not easy. The GMV Group is of the view that the industry has to consider that the safety requirements for self-driving vehicles are much more demanding than those in sectors such as civil aviation, with all the implications that this might have from a RAMS perspective. Thus, implementing higher levels of autonomy on new models, which rely on mature safety concepts, might take time.

The fast evolution of technology has to be coupled with automotive engineering cycles and safety concepts, which need to be present as the first driver in autonomous vehicles. On the other hand, there is the need for the vehicles to be prepared to be updated on a periodic basis, considering the huge amount of Lines of Code which run inside them and how this software might need to be updated on-the-air. This has to be assured in a secure way, guaranteeing that the safety concept of these systems is not altered at all. Implementing secure OTA updates of the in-vehicle software will have new versions of the vehicle with every new version of the technology and software of it.

It is also essential to consider the role played by other technologies and sensors (e.g. 5G, cameras, HD maps, etc.) as GNSS / INS will have to work in an integrated manner with them in order to cope with the highly demanding requirements of higher levels of autonomy in vehicles. So, is autonomous driving a distant dream, or will technology innovations make it a reality in the near future; and what are groups like GMV doing to bring self-driving cars on our streets? In an interview to Geospatial World, Sara Gutiérrez Lanza, Automotive Business Unit Director at GMV, answered some of these questions. Excerpts

Can you tell us about GMV and the group’s experience in the space sector?

Founded in 1984, GMV is trading globally in sectors such as aerospace, defense and security, cybersecurity, intelligent transportation systems and automotive. With nearly 2000 professionals, GMV has subsidiaries in Germany, Colombia, France, USA, Malaysia, Poland, Portugal, UK and Romania, and 65% of its turnover comes from international projects implemented across the globe.

In the space market, the areas of activity of GMV encompass global navigation satellite systems and applications (GNSS), flight dynamics, satellite control centers, mission analysis, mission planning and scheduling, payload data segments, science operations centers, user segments and space applications, satellite communications, simulation, systems engineering and GNC, and on-board and embedded software. 

GMV is an important player in the global GNSS community and is involved both in the development of GNSS infrastructure and the provision of services based on GNSS for different markets. Since 1988, when the European Space Agency awarded the group with a contract to develop GPS processing algorithms for precise satellite positioning, GMV has defined state-of-the art GNSS systems in Europe.

We have developed advanced techniques for precise orbit determination, time synchronization, integrity, user receiver algorithms, data pre-processing, Ionosphere, HW bias and troposphere estimation, navigation message generation, real time SW, HW infrastructure, and processing algorithms for ground-based augmentation systems (GBAS). The group has developed the core algorithmic facilities for both the flagship European GNSS Programs, EGNOS and Galileo.

What all in-house autonomous car solutions has the group devised?

In the automotive sector, GMV has been an active supplier for more than 10 years now, providing different hardware and embedded software solutions.

After years of experience, the capabilities of GMV in the sector and in particular, for the autonomous cars, focus around the following areas:       

• Automotive Platforms (both hardware and embedded SW)
• Connected Car Cloud services
• Advanced positioning algorithms and solutions for autonomous driving and ADAS applications 
• Cybersecurity solutions and services

Based on the experience gained in the development of integrity solutions for EGNOS and Galileo programs, GMV has been working for more than 10 years in developing GNSS integrity solutions for the automotive market (with 9 patents in the subject up to date). The group has also developed its own operational proprietary state-of-the-art GNSS precise point positioning (PPP) solution, known as magicPPP®, which is a key element of the company’s magicGNSS® (https://magicgnss.gmv.com/) suite of GNSS products and services.

This is the solution on top of which the offer to BMW is built, entailing both a positioning engine software which calculates the vehicle’s position and other magnitudes, using advanced GMV-developed algorithms, including components that have already been patented. These algorithms have been specially modified and adapted to meet BMW’s performance and safety requirements. The software abides by the most demanding automotive standards and the highest quality levels of safety. Another key component of the solution provided is a GNSS correction service to be run in a secure infrastructure using data from a global network of monitoring stations to be set up by GMV under this contract.

Can you please throw some light on GMV’s specific cybersecurity solutions for connected autonomous vehicles?

GMV offers a holistic approach to cybersecurity of CAV, which includes specialized services for vulnerability analysis (of systems, software, ECUs etc.), advanced penetration tests at different levels (from software modules to ECUs or the vehicle as a whole system), proactive services for cybersecurity assessment, antimalware laboratories, cyber-intelligence  services including the analysis of new cyber threats by constant monitoring of the dark / deep web, zero-day attack detection (based on behavior) and reactive services such as management of incidents (possibly involving GMV’s Computer Security Incident Response Team (GMV-CERT) or forensic analyses. On top of this, GMV is consolidating a cyber-secure car product suite which includes products such as IDPS (Intrusion Detection and Prevention Systems) or Secure Digital Key solutions.

A major issue with GNSS systems is that they lack integrity mechanisms. How can this problem be addressed?

Integrity is a key enabler of the use of GNSS in critical applications such as those present in autonomous driving. In addition to outstanding performances in terms of accuracy, convergence time or availability, the technology to be used must comply with highly demanding integrity requirements. GMV’s highly precise and safe positioning solution for autonomous driving is based on an integrity concept implemented consistently throughout the complete system and involving both the corrections service and positioning engine. The cornerstone for protection levels calculation in the positioning engine are several integrity algorithms patented by GMV, which have been duly adapted to work integrated within a PPP positioning algorithm.

We are applying a state-of-the-art functional safety methodology stemming from our unique experience in other sectors such as civil aviation which are aligned with functional safety approaches commonly applied in the automotive sector with the purpose to demonstrate and validate the highly demanding integrity requirements which are demanded in the applications the solution is targeting.

Also Read: Why LiDAR is important for autonomous vehicles?

Comments are closed.