Autonomous Vehicles to predict pedestrian movement

Will Autonomous Cars recognise drunks

pedestrians-umichUniversity of Michigan researchers are teaching self-driving cars to recognize and predict pedestrian movements.

Prior work in this area has typically only looked at still images.

Predictive power requires the network to look into the minutiae of human movement: the pace of a human’s gait (periodicity), the mirror symmetry of limbs, and the way in which foot placement affects stability during walking.

One researcher said “If a pedestrian is playing with their phone, you know they’re distracted” “Their pose and where they’re looking is telling you a lot about their level of attentiveness. It’s also telling you a lot about what they’re capable of doing next.”

We wonder if that will have a random generation movement factor for children or drunks.

About David Brown 521 Articles
David’s boyhood passion for motor cars did not immediately lead to a professional role in the motor industry. A qualified Civil Engineer he specialised in traffic engineering and transport planning. What followed were various positions including being seconded to a government think-tank for the planning of transport firstly in Sydney and then for the whole of NSW. After working with the NRMA and as a consultant he moved to being an independent writer and commentator on the broader areas of transport and the more specific areas of the cars we drive. His half hour motoring program “Overdrive” has been described as an “informed, humorous and irreverent look at motoring and transport from Australia and overseas”. It is heard on 22 stations across Australia. He does weekly interviews with several ABC radio stations and is also heard on commercial radio in Sydney. David has written for metropolitan and regional newspapers and has presented regular segments on metropolitan and regional television stations. David is also a contributor for AnyAuto