Near/in-sensor computing for edge AI
PhD - Antwerpen | Just now
Energy-efficient computing systems, such as neuromorphic ones, are emerging as compelling solutions for low-latency, extreme edge applications [1]. However, the deployment of truly robust and reliable systems of this type into real-world scenarios entails delivering solutions that can cope with at least two sources of variability: environmental (e.g., background illumination changes) and sensor variability (e.g., sensor drift). This becomes extremely crucial for use case scenarios involving multiple sensory nodes or hardly accessible experimental platforms, typically processing large amount of data and with limited output bandwidth (e.g., in automotive, underwater or space applications). A compelling solution to tackle this challenge is found in the emergent paradigm of AI-driven “near/in-sensor” computing, which has been shown to drastically reduce the bandwidth usage, leading to huge power saving and low latency response [2, 3]. While successful demonstrations of this paradigm have recently been presented for extreme edge biomedical applications [3,4], its potential in the neighboring automotive and underwater sensing domains is yet to be unleashed. Thus, the goal of this PhD project is to investigate how leveraging AI-driven near/in-sensor computing paradigms affects the sensing quality in automotive and underwater use cases. Initially, the focus will be on quantifying how augmenting the sensor transduction with an adaptive feedback loop can enhance the sensing quality in dynamically changing environments, as opposed to open loop systems. Next, the goal will be to validate such principles in simulation environments. This entails the identification of suitable benchmarks as well as the generation of dedicated datasets. A successful outcome of this research has the potential to deliver the next generation of truly intelligent, “self-calibrating” sensing systems for edge computing applications.
We offer you a challenging, stimulating, and pleasant research environment, where you can contribute to international research on artificial intelligence with a close link to the underlying hardware. While you will work in the AI&Algorithms department, you will also be working together with imec hardware, sensor development and university teams on jointly producing novel solutions.
Our ideal candidate for this position has the following skills:
You have a Master’s degree in Computer Science, Informatics, Physics, Engineering or Electronics
You have knowledge about artificial intelligence and machine learning
You have interest in algorithmic and system design
You have good programming skills and are flexible in the use of software and coding tools or libraries (git, pytorch, tensorflow, …)
Understanding of the physics and properties of different sensor modalities (lightwaves, radiowaves, etc.) is considered a plus
Experience working with neuromorphic hardware and software is considered a plus
You can plan and carry out your tasks in an independent way
You have strong analytical skills to interpret the obtained research results
You are a responsible, communicative, and flexible person
You are a team player
Your English is fluent, both speaking and writing
References:
[1] Corradi, Federico, et al. "Neuromorphic Edge Computing: Challenges, Opportunities, and Current Solutions." (2025).
[2] Zhou, Feichi, and Yang Chai. "Near-sensor and in-sensor computing." Nature Electronics 3.11 (2020): 664-671.
[3] Safa, Ali, et al. "Neuromorphic near-sensor computing: From event-based sensing to edge learning." Ieee Micro 42.6 (2022): 88-95.
[4] He, Yuming, et al. "An implantable neuromorphic sensing system featuring near-sensor computation and send-on-delta transmission for wireless neural sensing of peripheral nerves." IEEE Journal of Solid-State Circuits 57.10 (2022): 3058-3070.
Required background: Master’s degree in Computer Science, Informatics, Physics, Engineering or Electronics, with knowledge about artificial intelligence and machine learning
Type of work: 80% modeling/simulation, 20% literature
Supervisor: Roel Wuyts
Daily advisor: Julie Moeyersoms, Nicoletta Risi
The reference code for this position is 2026-121. Mention this reference code on your application form.