The Swedish Defence Research Agency, FOI, is a government agency conducting research within the fields of defence and security. FOI has recently taken an interest in neuromorphic computing because of the possibility of developing low-power and low-latency sensors. This interest is split between two categories: defense and security applications for event cameras; and neuromorphic computing for low-power sensor systems.
The work on event cameras covers a broad spectrum of applications such as object detection (e.g. UAVs), classification, depth estimation, navigation, object tracking, fluid dynamics (e.g. explosions), self-supervised learning, and optical communication. This work typically employs more traditional signal processing or machine learning with non-spiking neural networks.
The work on neuromorphic computing concerns the use of spiking neural networks, or other neuromorphic techniques, to develop low-power sensors. This work includes also other types of sensors such as microphones, radio, and radar. When possible, the aim is to evaluate algorithms on neuromorphic hardware.
If the above sounds interesting, then FOI regularly supervises projects in these research areas. Many projects are announced at the following link: https://www.foi.se/jobba-hos-oss/student-som-gillar-utmaningar-exjobba-hos-oss.html, however you can also contact Hannes Ovrén directly at hannes.ovren@foi.se. 

Note that Swedish citizenship is mandatory for students doing projects at FOI;
but some projects could be implemented at KTH with FOI being an external co-supervisor.