What do I want my autonomous vehicle to do?

Perceive objects and recognize them perfectly just like how human eyes do !
Analyse situations and take time actions accurately as the human brain does !!
Sense environment and adjust the conditions like how a human body does !!!

An autonomous driver with human intelligence is the ideal solution to solve few issues that the researchers are trying to fix to make it absolutely safe. Huge number of technologies such as artificial intelligence, neural networking, machine learning, best in class sensors – LIDARs, RADARs, mapping, rule based decision making are all being tried and tested to make a fully autonomous car hit the streets at the earliest. Big players in the automotive industry are banking on the most crucial technology called the “cognitive computing”.

So, isn’t cognitive computing similar to AI? Is Cognitive computing different to AI?

Umm.. well, no! Let’s agree that though AI helps machines to do intelligent things on their own, where the trained data and scenarios empower the machine’s decision making process, it can neither match the human instincts nor with the human intelligence . This limitation is now overcome by the use of cognitive computing, which mainly deals with the simulation of human brain. Cognitive computing can be called as the use of deep learning for real time applications.

IBM defines the end result of cognitive computing as “systems that learn at scale, reason with purpose and interact with humans naturally” which sounds like a cliche now but will soon be a reality!

The huge “trust” issue that we have on autonomous cars, whether or not we can confide in them completely can possibly be attempted to be resolved with cognitive computing. Quick adaptation to different situations and contexts with minimum / no human supervision and natural language interaction makes cognitive computing the go to technology for autonomous vehicles. Human brain imitation with the help of complex data analytics and pattern recognition algorithms is much needed for the auto-pilot or driver assistance systems to mindfully handle critical dynamic operations.

LSTM – Long short term memory, a DNN variant is said to use spatiotemporal context which can quite predict the future (I wish !!!). Several chip manufacturers, AV researchers, software developers are increasingly adopting cognitive computing in the recent days to achieve highly independent, L-5 autonomous vehicles with least chances of undesirable outcomes.

Who all are working on Cognitive computing?

Several startups have partnered with big player to bring cognitive computing to autonomous driving.

Five.ai  – Cognitive computing for perception, agent intention, motion planning, Safe vehicle operation

Cortica – Mainly works on decision making, situation awareness & predictability

Isee.ai – Next generation AI for autonomous vehicles

Argo.ai – Situational awareness and dynamic driving using advanced AI and robotics

Brain corp – Path planning and adaptive behaviour for self driving

Humanising autonomy – Measuring human intent and situation analysis

Deepmind – Unsupervised learning

Safe, fully autonomous cars are expected to hit the streets by the next decade and with all the human intelligence transferred to systems, we need to get ready to be guided by them as well.

Did you enjoy this article?
Signup today and receive free updates straight in your inbox. We will never share or sell your email address.
Liked this Blog? Then Please ShareTweet about this on Twitter
Share on Facebook
Share on LinkedIn
Email this to someone