CB @ Semicast Research9 Oct 2023 14:01
this chart summarizes the development of the global #automotive #dms market. the dms installation rate is forecast to rise from 4% in 2021 to 85% in 2030. in the course of a decade, dms will move from being a niche #technology which many have written-off as obsolete, to a #safety technology which saves as many lives as #airbags and #seatbelts. regulatory and advisory bodies have been paying close attention for several years, even though the effectiveness of dms technology to save lives has been drowned out by the #selfdriving developers and their lobbyists, comms, and pr people.
dms looks like a #computervision problem. but in practice dms is a technology to solve a people problem, which is that humans get distracted and drowsy, and some drive while impaired by alcohol or drugs. so effective dms means development of a complete system which sees and understands human behavior as it applies to crash risk. not every turn of the head away from the roadway represents distraction; not every eye closure represents drowsiness. critically, not all dms are created equal.
a low-cost system which beeps every time a person turns their head or closes their eyes and has no understanding of the subtlety and nuance of human behavior, becomes "super-annoying" super quickly. thus, #interiorsensing is giving way to interior #perception, and underpinning perception is the necessity of #humanfactors behavioral science.
the key challenge currently facing dms developers is the lack of harmonization across the us, europe, china and #ncap, of what a dms actually does. there looks to be agreement coming of a basic need for #drowsiness, #distraction, and #impairment detection, driven primarily by euro ncap - for safer cars, vans & trucks. on top of that lies a superset of features enabling automated driving and functions in the #digital****pit. euro ncap is constantly raising the minimum performance bar, with a requirement for cognitive distraction already required as early as 2029. a demonstrable technology roadmap and commitment to ongoing investment in #innovation is therefore critical to success in dms development.
#automakers, tier-1s and semiconductor and component suppliers must therefore choose their dms roadmaps extremely carefully. to support this need, semicast is adding significant additional segmentation to its analysis of the automotive dms market in its upcoming 2024 study. this is summarized as follows:
- by movement detection into "owl" and "lizard" detection
- by feature set
- by processor type
- by cmos image sensor location
- by cmos image sensor resolution
- by driver and occupant monitoring
- by front-seat and whole-cabin monitoring
message me for details of the 2024 edition of #semicast's report "#adas & automotive vision-based driver monitoring systems.” my messages are open. availability of the research is limited to 30 purchasing companies.
https://www.linkedin.com/feed/update/urn:li:activity:7117104083211010049/