Fullmoonfever
Top 20
Just having a look at BMWs Tech Radar page.
Neuromorphic listed as a watch in their radar. They def understand its potential
We've already discussed they testing / POC with SynSense for the Smart Cockpit / occupant behaviour (shoulda gone with NVISO / Akida imo haha).
I found it interesting though, as below, that their accoustic vehicle access / speech recognition is still listed as neuromorphic but they don't specify working with Synsense (or who actually) for that POC.
Be nice if they were taking a leaf out of MBs book
www.bmwgroup.com
Neuromorphic listed as a watch in their radar. They def understand its potential
We've already discussed they testing / POC with SynSense for the Smart Cockpit / occupant behaviour (shoulda gone with NVISO / Akida imo haha).
I found it interesting though, as below, that their accoustic vehicle access / speech recognition is still listed as neuromorphic but they don't specify working with Synsense (or who actually) for that POC.
Be nice if they were taking a leaf out of MBs book
Technology Trend Radar
The BMW Group Trend Radar provides deep insights into our strategy of finding the most innovative products and services for our customers.
WATCH
- 4. NEUROMORPHIC COMPUTING
NEUROMORPHIC COMPUTING IS INSPIRED BY THE NEURAL STRUCTURE OF THE HUMAN BRAIN.
Neuromorphic computing offers a complementary path next to the conventional architecture of traditional processors. Rising technologies like autonomous driving or natural language processing rely on the usage of big data and machine learning – both in need of much more powerful ways of computing. However, simulating networks of “neurons” with traditional silicon can be costly in terms of time, size, and energy consumption. In contrast, by imitating the human brain, with transistors physically operating as neurons, neuromorphic computing is a huge step up in adaptability, speed and energy efficiency for neural network-based AI. Using neuromorphic chips, it will also be possible in the future to integrate powerful AI directly into vehiclse to perform essential computational operations on-board.
CAPABILITIES- Higher energy efficiency of neuromorphic chips compared to conventional ones
- Enabling real-time signal processing
- Context aware chips due to low latency and seamless sensor fusion of camera, lidar, audio, etc.
- Enabling new paths to unsupervised learning
- Acoustic vehicle access: PoC for on-board speech recognition.
- Smart cockpit: POC with SynSense to develop neuromorphic chips in combination with cameras in order to track the behavior (e.g. attention) of occupants very efficiently.
- Segmentation and processing of images in realtime.