Research Topics 2021


Please review further details of the research topics below and click on the relevant link to apply. Remember that you can start your application and return to complete it any time before the deadline of March 15th 2021.


Topic 1: System Security for Infrastructure Control Systems

Background

Recently, industrial control systems (ICS) for critical infrastructure, such as power plants and water distribution control and management systems, have been digitalized. While the digitalization achieves sophistication of data driven O&M and optimization of the control algorithm, there are emerging cyber threats against industrial control system as well as vulnerability exploitation and sensitive data exfiltration. Many sophisticated attacks exploit zero-day vulnerabilities. Since exploitation of a vulnerability in such systems may cause severe consequences in the physical world, it is a heavy responsibility of system/device vendor including Toshiba to mitigate vulnerabilities and attack surface. While the mitigation can be done through vulnerability testing and exploit synthesis using binary program analysis[1][2], there are remaining challenges specific to ICS such as timing constraint consideration, non-standard instruction set and byte code interpreters. Furthermore, data utilization of the infrastructure control field also follows increasingly by progress of AI technology etc., and we expect growing needs for the data protection technology and security protocols to leak with a disclosure risk of sensitive information.[3][4].

Suggested Research Proposals

  • Security Verification for Software Implementation in industrial control systems and devices, in detail, finding vulnerabilities, composing attacks and/or synthesizing patches technology for binary programs and systems. There have already been many studies in this field, so it is important to extend existing techniques with new idea to apply them to industrial control system and devices.
  • Data protection technology and data protection protocols suitable for industrial control system, such as, on the premise of multiple systems, a secure search of various O&M data and a secure fault prediction/detection technique of the system, furthermore industrial data analysis without exposing sensitive information, such as searchable encryption.

Related Papers

[1] T. Avgerinos, et al. AEG: Automatic Exploit Generation. In Proc. of NDSS, 2011.

[2] P. Godefroid, et al. Automated Whitebox Fuzz Testing. In Proc. of NDSS, 2008.

[3] Komano, et al., Toward Highly Secure Metering Data Management in the Smart Grid, https://www.imi.kyushu-u.ac.jp/files/imipublishattachment/file/math_58d8ad2f89418.pdf

[4] Start of Project to Verify Open Platform Aggregation Business, https://www.toshiba-energy.com/en/info/info2020_0608.htm

APPLY HERE

 


Topic 2: Advanced Digital Healthcare, Digital Twin of People

Background

Due to the impact of COVID-19, non-contact vital measurement technology to ensure social distance is important. Non-contact vital measurement technology is known to have low measurement accuracy and unstable readings due to external influences, such as brightness in the captured images, instead of low subject load.

Data and image-based health measurement technology [1][2] and human digital twin technology will be important with the new normal.

AI technology, smartphones and cloud computing are becoming more popular. This has created an environment for the advancement of digital health.

At Toshiba, Research and Development Center and Toshiba Digital Solutions provide disease prediction services for SOMPO Holdings. Toshiba is also using technology from an Israel start-up to measure health and stress levels using smartphone cameras, and early detection of dementia technology from Silicon Valley to expand its services to insurance companies in Japan.

Toshiba Group holds smart receipt purchase data, which can be used to build a more detailed digital twin of people. The digital twin of people is expected to enable new services such as content recommendations and optimization of social behavior. Synergies are also expected through collaboration with data obtained from the precision medicine business, such as micro RNA [3][4] analysis service.

Suggested Research Proposals

  • Technology that measures a person’s health, mental state and behavior using low-burden methods such as images from a smartphone camera and records the results with high accuracy and high robustness using machine learning and other techniques.
  • Digital technology to create new services which will sense information about people.

Related Papers

[1] Daniel McDuff, Christophe Hurter, and Mar Gonzalez-Franco. 2017. Pulse and vital sign measurement in mixed reality using a HoloLens. In Proceedings of the 23rd ACM Symposium on Virtual Reality Software and Technology (VRST '17). Association for Computing Machinery, New York, NY, USA, Article 34, 1–9.
https://www.microsoft.com/en-us/research/wp-content/uploads/2017/10/a34-mcduff.pdf

[2] Fadel Adib, Hongzi Mao, Zachary Kabelac, Dina Katabi, and Robert C. Miller. 2015. Smart Homes that Monitor Breathing and Heart Rate. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (CHI '15). Association for Computing Machinery, New York, NY, USA, 837–846. http://people.csail.mit.edu/fadel/papers/vitalradio-paper.pdf

[3] Toshiba Research News: https://www.toshiba.co.jp/rdc/rd/detail_e/e1911_06.html

[4] Toshiba Clip: https://www.toshiba-clip.com/en/detail/p=641

APPLY HERE

 


Topic 3: Research & Development of Information Acquisition and Colloboration Infrastuctures for Smart Cities

Background

IoT technologies such as the cloud, wide-area communication network, and so on have made it possible to collect data of infrastructures such as energy, communication, transportation, crime, and disaster information.

However, at present, these data are still only used in their own domains, for example, power consumption data is limited to only power companies. Meaning a "smart service" which links data from one domain to another in a different organizations has not yet been created (e.g., between companies).

In order to invent resilient smart city functions, which lead to disaster prevention / mitigation and sustainable smart city functions and which lead to de-carbonization. It is necessary to link data from these various domains efficiently and securely to analysis and control. The development of the infrastructure (platforms, data formats, etc.) to link the information has become an issue. More specifically, development of secure data distribution method, ensuring scalability for large and wide variety data, and development of standardized APIs for data sharing, etc.

Toshiba have businesses in the multiple infrastructure domains such as energy, building facility, industrial equipment, and so on. There is an opportunity for us to contribute to further smartening of society by utilizing data from these areas.

Suggested Research Proposals

  • Research on platform technology for data utilization between smart city infrastructures, for examples, system linkage, middleware, APIs, for System of Systems (SoS) and infrastructure services that utilize this.
  • Research on data control technology (flow control, integrity, authenticity, traceability, etc.) for securely linking and distributing data across various domains (industrial areas, regions, processes, etc.).

Related Papers

[1] ITU-T Y.4201 : High-level requirements and reference framework of smart city platforms
[2] Larissa Romualdo-Suzuki and Anthony Finkelstein, "Data as Infrastructure for Smart Cities: Linking Data Platforms to Business Strategies", 2020, arXiv:2005.11414 [cs.CY]

APPLY HERE

 


Topic 4: Realizing a Safe and Secure City: Next Generation Atmospheric and Hydrospheric Sensing Technologies

Background

Natural disasters appear to be increasing in number each passing year, causing not only economic damage but also mental and physical hardship for those affected. In order to protect cities and its citizens from natural disasters, it is necessary to detect various atmospheric and hydrospheric occurrences as well as observe cumulonimbus clouds with the aid of weather radar. Even with a supercomputer it is difficult to accurately predict when and where heavy rainfall and strong winds will occur. Toshiba has developed a phased array weather radar, which enables forecasting of up to 30 minutes before heavy rainfall [1][2]. However, in order to help more people evacuate, it is necessary to capture the signs of cloud condensation hours earlier. Although accurate forecasting is beneficial for preemptive evacuation an aerial technology that detects in real-time rivers breaking their banks, residential flooding and tornado movement would be an excellent tool for rescue services as well as for citizens who have yet to evacuate. In addition, it would make it easier to calculate the economic damage ensuring quick and efficient reconstruction. In this research, 3D real-time sensing technology that observes the wind flow and water vapor movement in the sky as well as 2D real-time sensing technology that captures tornado and flooding on the ground are required. In addition, sensor fusion that combines multiple sensing technologies is expected to play a crucial part in producing a safe and secure city.

Suggested Research Proposals

  • Research and development of ground-based radar and radiometer for real-time 3D sensing of water vapor
  • Research and development of Doppler LIDAR, etc. for real-time 3D sensing of wind high up in the atmosphere as well as near ground level
  • Research and development of sensors for real-time 2D sensing of hydrospheric phenomena
  • Research and development of sensor fusion technology that combines multiple meteorological and hydrospheric sensing technologies

Related Papers

[1] https://www.toshiba-clip.com/en/detail/5826

[2] https://library.wmo.int/pmb_ged/iom_109_en/Session5/O5_02_Mizutani_Active_Phased_Array_Weather_Radar.pdf

APPLY HERE

 


Topic 5: Reliable AI Contributing to Trustworthy Infrastructure Services

Background

AI technologies are remarkably advanced due to development of deep learning, so it is certain that AI will act some of the operations and maintenance of infrastructure such as electricity, transportation, and logistics, which are currently performed by humans. In Japan, the production workforce in 2030 is expected to decrease by 10% compared to 2018, and improving the efficiency of infrastructure O&M is a significant issue for a future sustainable society. Since this trend is also a big business opportunity, Toshiba has announced moves toward further growth as an infrastructure services company.

Toshiba provides various operation and/or maintenance services for infrastructures such as power generation, water and sewage, highways and railways and contributes to developing the related industries and improving quality of life. These infrastructures play a very important role in human society, so the introduction of AI cannot allow them to become unstable. Therefore, to let AI act an important and responsible role, it does not only provide high performance for specific tasks but also obtain reliability for humans by combining explainability, security and fairness. These themes are getting big attention at academic societies but further research is expected at the view of industrial applications.

Suggested Research Proposals

Research proposals could approach this from each viewpoint of explainability, security and fairness or an effort that integrates them. The points from each viewpoint are as follows.

  • Explainability: It is one of the hot topics on AI research. Though there are many studies that simply visualise the relationship between features and models, it is unclear whether they really "explain". We expect an industrially convincing explanation.
  • Security: Understanding defense against typical attacks on AI such as causing malfunction during inference by unexpected inputs, polluting the model by mixing malicious learning data.
  • Fairness: How to overcome data bias in machine learning and eliminate or suppress unfair discrimination (racism/ethnic, gender, cultural/regional discrimination, etc.).

Related Papers

[1] TOSHIBA Cyber Physical Systems: https://www.toshiba.co.jp/iot/index_en.htm
[2] Toyota Vehicles with Toshiba’s Advanced Image Recognition Processor Win Japan’s Highest Preventive Safety Performance Awards for Second Year https://toshiba.semicon-storage.com/ap-en/company/news/news-topics/2020/06/automotive-20200612-1.html
[3] Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey, arXiv:2006.11371, 2020.
[4] A Survey on Bias and Fairness in Machine Learning, arXiv:1908.09635, 2019.

APPLY HERE

 


Topic 6: Super accelerating scientific simulation with physical model embedded machine learning method for real-time digital-twin in cyber physical system.

Background

Real-time digital-twin concepts as an inter-mediator taking care of the interaction among human, nature and infrastructure systems would be to simulate the complex real world interactions as well as to run the real world in the cyber physical systems [1][2].
Digital-twin computing could well improve infrastructure resilience for various risks, and to optimize design, operation and maintenance in cyber physical systems.
Scientific simulations such as fluid mechanics, non-linear mechanics, multi-physics and multi-scale phenomena are implementations of theoretical models, they may run very slowly, even on modern computer hardware; for example, global climate models may take thousands of CPU-hours.
It is necessary to highly accelerate simulations --computer implementations of mathematical models—to test out new concepts and ideas, and explore potential behaviors of infrastructure systems and natural systems [3][4].

Toshiba has owned the data, model and know-how of infrastructure systems, and has potential and environment to develop digital-twin computing technologies [1][2].

Suggested Research Proposals

  • Accelerating finite-difference or finite-element simulation on Eulerian frame with machine learning method such as Lagrangian neural networks, Hamiltonian neural networks, deep neural networks, and convolutional networks [5][6].
  • Physical modeling method embedded machine learning for continuum mechanics or non-liner dynamical simulation based on physical law exploration or reduced order modeling method such as proper orthogonal decomposition, Galerkin projection, and sparse identification of nonlinear dynamical systems [7].
  • Accelerating hybrid simulations of multi-agent simulation and physical simulation for a large number of risk scenarios in infrastructure systems [8].
  • Making simulation based high-dimensional data predictable method such as randomly distributed embedding [9].

Related Papers

[1] https://www.toshiba.co.jp/iot/index_en.htm
[2] https://www.toshiba-energy.com/en/transmission/product/iot.htm
[3] https://www.infoq.com/news/2020/03/deep-learning-simulation/
[4] https://arxiv.org/abs/2001.0805
[5] https://arxiv.org/abs/2003.04630
[6] https://arxiv.org/abs/1906.01563
[7] https://arxiv.org/abs/1701.03424
[8] https://arxiv.org/abs/2004.05185
[9] https://www.pnas.org/content/115/43/E9994

APPLY HERE

 


Topic 7: Next Generation Innovative Electronic Devices

Background

Toshiba aims to be a CPS (Cyber Physical System) technology company that creates and provides new services and solutions using CPS technology. We are working on power semiconductor device technology, sensing device technology, and information storage device technology as technologies that support CPS technology, and in the future, we will promote further high functionality and high performance by integrating with IoT and AI technology. Specifically, we aim to create the following innovative next-generation devices and module technologies.

1) Next-generation high-performance power semiconductor device technology [1,2] and innovative intelligent power semiconductor module technology using Si, SiC, and GaN.
2) Innovative sensing device / module technology based on semiconductor technology such as MEMS [3,4].
3) Development of next-generation information storage device and spintronic device / module technology by integrating spintronic technology with artificial intelligence and machine learning[5,6].

Suggested Research Proposals

The following are examples of research themes that we expect as research proposals.

  • Research and development of innovative power semiconductor devices using Si, SiC and GaN that realize power saving.
  • Research on an approach for improving the performance of power semiconductor devices using AI technology.
  • Development of intelligent power semiconductor modules using advanced digital circuit technology.
  • Research on device control technology for inertial sensors and sensor fusion technology for position estimation.
  • Research on spintronic device design and data processing application technology for ultra-large capacity HDDs.
  • Research on device analysis and CPS module technology for innovative and ultra-sensitive spintronic sensors.

Not limited to examples described here, we look forward to receiving research proposals for future technologies related to innovative next-generation devices and module technologies in CPS fields.

Related Papers

[1] S. Asaba, et.al, "Breakthrough in Channel Mobility Limit of Nitrided Gate Insulator for SiC DMOSFET with Novel High-temperature N2 Annealing", in Proc of the 31st Int. Symp. on Power Semicond. Devices & ICs, p.139(2019)
[2] Y. Kajiwara, et.al, "Highly Reliable GaN-MOSFETs with High Channel Mobility Gate by Selective-Area Crystallization", in Proc of the 32nd Int. Symp. On Power Semicond. Devices & Ics, p.302(2020)
[3] R. Gando, "A Compact Microcontroller-based MEMS Rate Integrating Gyroscope Module with Automatic Asymmetry Calibration", in Proc. IEEE 33th Int. Conf. Micro Elect. Mech. Syst., p1296(2020)
[4] R. Gando, "A MEMS Rate Integrating Gyroscope Based On Catch-and-release Mechanism for Low-noise Continuous Angle Measurement", in Proc. IEEE 31st Int. Conf. Micro Elect. Mech. Syst., p.944(2018)
[5] T. Maeda, et.al, "Microwave Assisted Magnetic Recording Technology for HDDs Achieving Higher Recording Density", TOSHIBA REVIEW 74(6) p17(2019)
[6] S. Shirotori, et.al, "Symmetric response magnetoresistance sensor with low 1/f noise by using an anti-phase AC modulation bridge", in IEEE Transaction on Magnetics 3012655 (2020)

APPLY HERE

 


Topic 8: Automated Software Development

Background

Toshiba's Cyber-Physical Systems (CPS)[1] aim to provide sustainable and evolving civil infrastructure systems[2] and services. From the perspective of software development, we need to satisfy functionality and business requirements quickly[3]. For that reason, we often develop software and services that reuse existing assets such as open-source software (OSS). Additionally, we often need to modify or optimize the software in these systems and services based on the feedback obtained during operation and maintenance.

This research focuses on software development automation methods that could dramatically improve software development and maintenance efficiency. That includes automation of all stages within the software development process, from the selection of OSS at the new service planning stage to the development, test, operation, and maintenance stages.

If the research topic proposal relates to OSS, the successful candidate will likely need to collaborate with other companies in the Civil Infrastructure Platform (CIP) projects led by Toshiba. In addition, when the research topic relates to software quality, we aim to establish an automation method to meet the software quality standards for civil infrastructure systems and services in collaboration with business units.

Suggested Reseearch Proposals

At each stage of the software development process, we expect your proposals on any of the followings for automated software development.

  • Method for automated open source software selection based on products requirement
  • Method for automated software quality determination for testing
  • Method for automated test case generation for quality assurance and to satisfy software quality standards
  • Method for automated defect identification and source code correction from error logs, etc.
  • Method for automated regression test generation from source code changes

Related Papers

[1] TOSHIBA Cyber Physical Systems https://www.toshiba.co.jp/iot/index_en.htm
[2] Civil Infrastructure Platform (CIP) https://www.cip-project.org/
[3] CIP Software Update Workgroup https://wiki.linuxfoundation.org/civilinfrastructureplatform/cip-sw-updates

APPLY HERE