Automotive intelligence moves forwards with ‘Liquid AI’
By Dr. Tim Sandle
September 19, 2024
DIGITAL JOURNAL
An 'Apollo Go' autonomous taxi on a street in Beijing - Copyright AFP Jade GAO
Is a new era of automotive intelligence about to begin? This is the claim of Autobrains Technologies who are working on ‘Liquid AI’, a self-driving car technology. This approach is designed to solve some of the current autonomous driving challenges.
In addition, the technology seeks to enhance vehicle autonomy by dynamically adapting to complex driving environments. This adaptability is considered as essential to achieving smarter and safer automotive solutions.
Such challenges include:
Edge Cases
An edge case is a problem or situation that occurs only at an extreme (maximum or minimum) operating parameter.
The infinite variety of unexpected driving scenarios presents conventional AIs with practically unsolvable tasks. Attempts to address this by feeding the systems more labelled images result in a loss of trackability and controllability.
Cost
Addressing real-world driving problems by expanding existing systems with more data, labelling, layers, and computational resources leads to escalating costs and power consumption.
Achieving a substantial improvement in system accuracy by a factor of 10 requires 10,000 times more computational resources.
Perception-Decision Disconnect
The missing interplay between perception and decision functions hinders effective and precise decision-making. For the AI to make optimal driving choices, it requires specific information. However, when details are missing or overly complex, precision is compromised, leading to incorrect reactions.
Liquid AI – Human Brain-Inspired
The technology combines Autobrains’ signature-based self-learning approach with a modular and adaptive architecture of specialized, scenario-based end-to-end skills.
According to Autobrains’ Founder and CEO, Igal Raichelgauz: “While current technologies perform well in handling average conventional driving tasks, they fall short when faced with unexpected real-world driving scenarios that demand greater precision. By using or implementing our Liquid AI, automotive companies can close their AI gaps”.
Autobrains draws inspiration from the human brain. As the human brain adapts its architecture based on context – such as light/weather conditions, surroundings, and relevant road users – Liquid AI has been designed to follow the same approach.
The basis of the technology includes:
Network of Specialized Narrow AI
Liquid AI comprises hundreds of thousands of specialized narrow AIs, each designed for specific tasks, making reactions very precise and tailored to the relevant driving scenario.
This specialized AI approach enables scalability, ranging from a few tens to hundreds of AIs for ADAS systems, scaling up to thousands for higher levels of automated driving, all the way to hundreds of thousands of AIs for full self-driving.
Adaptive Architecture
Unlike fixed systems, Liquid AI’s architecture adapts dynamically to the driving context, activating only relevant modules as necessary. This significantly reduces power consumption and compute requirements, not only resulting in cost savings for the System on Chip (SoC) hardware.
Efficiency and Precision
By mimicking the brain’s flexibility, Liquid AI achieves superior performance, cost-effectiveness, and safety. This includes human-like cognitive processing, which mimics human decision-making, allowing for better handling of unpredictable real-world conditions.
Efficient Resource Utilization
Lower computational power requirements make it scalable across various vehicle models without compromising performance.
These factors lead to a potentialenhancement in situational awareness and decision-making, providing a safer and more reliable driving experience.
Read more: https://www.digitaljournal.com/tech-science/automotive-intelligence-moves-forwards-with-liquid-ai/article#ixzz8mVkwNvN4
An 'Apollo Go' autonomous taxi on a street in Beijing - Copyright AFP Jade GAO
Is a new era of automotive intelligence about to begin? This is the claim of Autobrains Technologies who are working on ‘Liquid AI’, a self-driving car technology. This approach is designed to solve some of the current autonomous driving challenges.
In addition, the technology seeks to enhance vehicle autonomy by dynamically adapting to complex driving environments. This adaptability is considered as essential to achieving smarter and safer automotive solutions.
Such challenges include:
Edge Cases
An edge case is a problem or situation that occurs only at an extreme (maximum or minimum) operating parameter.
The infinite variety of unexpected driving scenarios presents conventional AIs with practically unsolvable tasks. Attempts to address this by feeding the systems more labelled images result in a loss of trackability and controllability.
Cost
Addressing real-world driving problems by expanding existing systems with more data, labelling, layers, and computational resources leads to escalating costs and power consumption.
Achieving a substantial improvement in system accuracy by a factor of 10 requires 10,000 times more computational resources.
Perception-Decision Disconnect
The missing interplay between perception and decision functions hinders effective and precise decision-making. For the AI to make optimal driving choices, it requires specific information. However, when details are missing or overly complex, precision is compromised, leading to incorrect reactions.
Liquid AI – Human Brain-Inspired
The technology combines Autobrains’ signature-based self-learning approach with a modular and adaptive architecture of specialized, scenario-based end-to-end skills.
According to Autobrains’ Founder and CEO, Igal Raichelgauz: “While current technologies perform well in handling average conventional driving tasks, they fall short when faced with unexpected real-world driving scenarios that demand greater precision. By using or implementing our Liquid AI, automotive companies can close their AI gaps”.
Autobrains draws inspiration from the human brain. As the human brain adapts its architecture based on context – such as light/weather conditions, surroundings, and relevant road users – Liquid AI has been designed to follow the same approach.
The basis of the technology includes:
Network of Specialized Narrow AI
Liquid AI comprises hundreds of thousands of specialized narrow AIs, each designed for specific tasks, making reactions very precise and tailored to the relevant driving scenario.
This specialized AI approach enables scalability, ranging from a few tens to hundreds of AIs for ADAS systems, scaling up to thousands for higher levels of automated driving, all the way to hundreds of thousands of AIs for full self-driving.
Adaptive Architecture
Unlike fixed systems, Liquid AI’s architecture adapts dynamically to the driving context, activating only relevant modules as necessary. This significantly reduces power consumption and compute requirements, not only resulting in cost savings for the System on Chip (SoC) hardware.
Efficiency and Precision
By mimicking the brain’s flexibility, Liquid AI achieves superior performance, cost-effectiveness, and safety. This includes human-like cognitive processing, which mimics human decision-making, allowing for better handling of unpredictable real-world conditions.
Efficient Resource Utilization
Lower computational power requirements make it scalable across various vehicle models without compromising performance.
These factors lead to a potentialenhancement in situational awareness and decision-making, providing a safer and more reliable driving experience.
Read more: https://www.digitaljournal.com/tech-science/automotive-intelligence-moves-forwards-with-liquid-ai/article#ixzz8mVkwNvN4
No comments:
Post a Comment