Overcoming roadblocks in tackling drowsy driving
We catch up with Mansoor Nasir, Senior Research Engineer with Mobis North America. Mansoor discusses his latest research on physiological monitoring, the development of driver monitoring systems, and the impact of covid-19 on the road to full autonomy. Mansoor will be presenting on the 16th September at our AutoSens Brussels edition on ‘From Driver Monitoring to Driver Engagement’.
You are a subject matter expert in Driver monitoring system, feature extraction and advanced data analytics at Hyundai-Mobis Technical Center, what do you bring in from your research on physiological monitoring into this role?
With the amount of time everyone spends in their vehicles, there is a great opportunity to perform long term non-intrusive physiological monitoring and get an insight into overall health of the drivers and passengers. One barrier for inclusion of physiological monitoring has been the lack of people with expertise in MedTech in the automotive world. In my opinion, the necessary tools already exist and with the inclusion of people with diverse backgrounds like mine, this area can become active for new research and development. In turn, MedTech will benefit from the cost reduction, streamlined manufacturing processes and focus on user-centric design that are established principles in the automotive world.
Mobis Technical Center of North America (MTCA) and Virginia Tech Transport Institute (VTTI) collaborated to gather to create a unique drowsy driving database, can you share more about this collaboration?
I will cover some details in my talk but essentially, one major roadblock in tackling drowsy driving has been the access data from such instances as they occur so infrequently. Conducting a real driving study, designed to cause higher instances of drowsy driving, gave us insights into driver inattention behaviour and allowed MTCA to build a Driver monitoring system (DMS) that can accurately detect most instances of intermediate and extreme drowsiness. Since the study included men and women from different age and ethnic groups, the data helped us to personalize the DMS system to each driver. This helps us reduce false warnings and further improve detection accuracy.
What are the most important features of the development process next generation HMI and specifically driver monitoring systems?
An effective DMS must include a driver engagement scheme that is supported by an intelligent HMI system. Until now, DMS systems have primarily focused on the detection aspect. Unfortunately, the warning systems are simplistic and ineffective in persuading an inattentive driver to take corrective action. By focusing on driver safety as a central paradigm and understanding driver psychology, next generation HMI systems can seamlessly and naturalistically provide vital information to the driver. These include features such as warnings escalation schemes, voice alerts, augmented reality displays, directional audio inputs and cabin environment changes to name a few.
Apart from DMS, which will play an integral part in the path to full autonomy, what do you think is the other key feature that the industry needs to develop a more robust version of?
OEMs and suppliers in the automotive world have to convince the consumers about the safety of autonomous driving technology. While everyone wants to reach the full autonomy milestone first, it is prudent to ensure that ADAS is not comprising driver safety. I, therefore, think that backup safety systems in all aspects of ADAS are extremely important so that in case of the failure of the primary system, the driver has enough time or warning to safely take over the control of the vehicle. The challenge is to be able to do so without making the vehicles economically unviable.
What impact on timescales to full autonomy do you think COVID-19 will have for automotive?
In the interim, it has been a setback for many companies due to manufacturing delays and also because ADAS systems by design are complex and require cross-functional teams to develop, test and validate. In the long run, however, it may actually fuel the interest and the urgency for this technology. A functional society which is not drastically affected by supply chain and delivery issues seen during the COVID crisis, is only possible with robot and autonomous services, whether for moving goods, food delivery or disinfecting public spaces. Thought-leaders in automotive world must utilize the unique circumstances created by COVID to envision how autonomous technology can alleviate the hardship faced by the masses in the unfortunate case of a similar occurrence in the future.
What are you looking forward to about presenting at AutoSens 2020?
It is unfortunate that our current circumstances will not allow me to attend AutoSens in-person but the conference speakers include folks from both research and industry. So I hope to network with other experts, exchange ideas and learn about the new tech. In future, when the conditions permit, it will be great to meet these people in-person as well.
Read latest news and updates from AutoSens
- Driving ADAS technology and autonomous vehicles into the future with Siemens ~ 10 November 20
- Wired Interfaces: Enabling End-to-End Interfaces for ADAS, Infotainment and Autonomy ~ 27 October 20
- Interview with Ross Jatou of ON Semiconductor and Junko Yoshida of EE Times ~ 9 October 20