As you have seen in previous automotive blogs by my colleagues, autonomous vehicles are a hot topic and Micron is an active partner in making this technology of the future a possibility. We have shown you the various steps towards autonomous driving, and the systems in the vehicle that come together to make it possible. I’d like to go into further detail on one of the Advanced Driver Assistance Systems (ADAS) applications, collision avoidance. In 2015, the National Transportation Safety Board (NTSB) recommended that forward collision avoidance systems become standard on all new US passenger and commercial vehicles. Collision avoidance systems are a growing part of advanced driver assist systems (ADAS) and the vehicle systems that enable Vehicle-to-Vehicle (V2V) communication. These systems include the “see and say” semiconductors that provide braking, parking and steering functions, such as:
- Adaptive Cruise Control: an optional cruise control system for road vehicles that is designated to automatically adjust the vehicle speed to maintain a safe distance from vehicles ahead.
- High Beam: headlight control strategy meant to continuously, automatically tailor the headlamp range so the beam just reaches other vehicles ahead, thus assisting in ensuring maximum possible seeing range without glaring other road users.
- Backup Cameras: attached to the rear of a vehicle to aid in backing up, and potentially to alleviate some rear blind spot.
- Automotive Night Vision: uses a thermographic camera aiming at increasing a driver's perception and seeing distance in darkness or poor weather beyond the reach of the vehicle's headlights.
- Parking Assistance/Automatic Parking: assists drivers in parking their vehicle.
- Collision Avoidance: system designed to alert and potentially reduce the severity of an accident.
- Traffic Sign Recognition: a technology by which a vehicle would be able to recognize the traffic signs put on the road e.g. "speed limit" or "children" or "turn ahead."
- Lane Departure Sensor: a mechanism designed to warn a driver when the vehicle begins to move out of its lane.
As collision avoidance systems become increasingly prevalent in vehicles as a result of consumer demand and government regulation, the population of cameras and LiDAR, radar and ultrasonic sensors proliferate and the need for memory increases. Micron predicts, an autonomous car will deploy up to 20 sensors containing camera, IR, LiDAR, radar, ultrasonic and other wireless sensor technologies. Self-driving cars will require fusion of these sensing inputs, generating GBs of data that will need to be consolidated within memory and storage accessible to computers capable of much greater performance than typical ECUs available today. As an example, LiDAR is used to collect data for creating high-resolution 3D digital images for collision avoidance. In a recent announcement, a vendor claims its LiDAR solutions are capable of producing 300k to 2.2 M data points per second with a range up to 200 m at cm-level accuracy.
There are four aspects to consider towards optimizing the memory capabilities in these systems.
- Instant-on: immediate availability of ADAS features such as back-up cameras, emergency brake system and blind spot detection.
- Speed: Latency to manage extremely high throughput of all the data coming from the sensors (currently 50 GB/s and projected to exceed 200 GB/s in future generations—2022).
- Safety and Security: how to protect mission critical systems and data from unauthorized access.
- Ultra-high temperature: operate at extreme temperature due to reduced physical space, smaller form factors, combined with higher performance.
Although technology is available that is used in other applications- such as consumer electronics, home automation and Internet of Things, implementation of these technologies into vehicles needs to be held to much higher standards.
Again, we assert that automotive memory requirements are extremely important and include:
- Quality first striving towards zero defects for safety-critical applications.
- Compliance to automotive industry and infrastructure standards which continue to evolve.
- Operational excellence with a robust supply chain that can rapidly adapt to demand fluctuations and provide continuous improvement.
Micron is the leading memory provider to the automotive industry (source: Gartner 2015), visit our Automotive Solutions page to learn more about our commitment to this industry.