Customer: Japanese Automotive Tier-1
Scope: Customer was looking for a partner to develop an Adaptive Driving Beam system integrating DLP to reduce glare on pedestrians and cyclists and to ensure more precise control of light distribution through independent LED control.
Solution:
- Platform: TI TDA 2X, DLP5531-Q1, Linux
- Conceptualization and competitive market analysis
- Selection of hardware platform based on performance analysis and peripheral interfacing
- Porting, optimization and development of pedestrian detection algorithm
- Deep learning models developed using Keras library
- Developed Python scripts for post processing
- Performance optimization and validation
- System integration
Impact:
Leveraging Sasken’s expertise, customer was one of the first to market with a solution that detects humans on the road and adjusts headlights such that there is zero glare on the pedestrian.
