Job Summary

Project: Enhanced Resiliency For ERP2 System

Duration: 3 years

Description: Next generation urban road usage charging systems utilize GNSS based vehicle detection. Project aims to develop system resiliency against GNSS signal disruptions to ensure robust and accurate vehicle positioning for charging.  The project focuses on development of data driven tools to support planning and management of new and existing charging points. Another objective is to design a cost-effective framework to detect malicious activities including jamming, blocking and spoofing of the GNSS signals. Furthermore, it aims to achieve robust and accurate vehicle positioning by leveraging on existing sensors and AI techniques to improve positioning in GNSS challenged environments. Additionally, the project explores the use of cost-effective beacons to enable V2X communication to enhance system resilience and accuracy.

Required Profile for Research Scientist

• Strong software development skills in C++ and Python. Hands-on experience with system design, modular architecture, and interfacing between multiple components.
• Familiarity with AI-assisted development workflows, including the use of AI tools for software development lifecycle (SDLC) automation to accelerate solution development.
• Ability to engage with stakeholders to formulate a comprehensive list of use cases and system requirements. Experience in system architecture design for integrated software solutions for the research outcomes.
• Expertise in data analytics and signal processing algorithms, with demonstrated ability to apply machine learning and AI techniques to real word problems.
• Experience in large-scale data mining and spatio-temporal analysis of raw GNSS data from road users to characterize signal quality and positioning performance across the road network. Derive high quality road segments by analysing attributes like DOP, SNR, and snapping distances to map links. Develop algorithms for charging point design, selection and quality assessment. Conduct root cause analysis for low accuracy charging points.
• Develop algorithm to detect jamming, spoofing, and blocking for GNSS data using hierarchical fusion of different sensors data collected from a vehicle OBU. Tracking algorithm to positively identify if users are driving on the roads and possibly engaging in malicious activities.
• Develop INS based dead reckoning algorithm using Kalman Filter variants. Design methods for extraction of map features and blind search to identify road segments based on INS-only data.
• Work on AI based multimodal data fusion and tracking algorithms for robust, high accuracy, real-time vehicle positioning in GNSS-challenged urban environments using multimodal data including GNSS, INS, V2X and others.

• Design test cases and conduct field trials to validate the algorithm performance for detection of malicious activities and improved vehicle positioning including problematic sites like carparks, urban canyon, tunnel, etc.