AI smart sensors for the commercial trucking and transportation industry connected with 5G - ON-820Project type: Innovation
Desired discipline(s): Engineering - computer / electrical, Engineering, Computer science, Mathematical Sciences, Physics / Astronomy, Natural Sciences
Company: Quantuity Analytics Inc
Project Length: 6 months to 1 year
Preferred start date: As soon as possible.
Language requirement: English
Location(s): ON, Canada; Canada; Canada; Canada
No. of positions: 1
Desired education level: Master's
Open to applicants registered at an institution outside of Canada: Yes
About the company:
Quantuity Analytics is developing an AI smart sensor to measure the air chamber push-rod stroke length in commercial trucks, saving the driver from visual measurement under the vehicle. Their sensor exceeds regulatory requirements, increases vehicle and driver safety, prevents unplanned maintenance issues, and saves 30 minutes of a commercial vehicle pre-trip inspection.
Describe the project.:
The project aims at implementing Tensorflow machine learning models deployed onto embedded SoCs (constrained edge devices), to perform machine learning inference using connected MEMS sensors on the physical properties of the environment they're installed in.
The solution is to be used in vehicle safety and compliance systems using intelligent IoT smart sensors with TinyML model deployed on each one to perform machine learning inference at the edge on every sensor and IoT gateway device. The project is to deliver a TRL7 product for first of 2 initial customers to validate its performance and functionality and complete the customer product validation step. We look to include a 5G IoT telematics system gateway to connect every vehicle so they will have the latest next-generation hardware and software tools to increase fleet operations performance, safety, reliability, and management from almost anywhere they operate their fleet services in the commercial trucking and transportation industry.
Our innovation is under a provisional patent filed through our attorney for a push-rod stroke length measurement to be performed using an AI model deployed on an embedded SoC wireless device. We use MEMS sensor data to collect the physical properties of a commercial air brake’s behaviour and performance under normal operating conditions, followed by train, test, validate, and deploy a pre-trained model onto our AI smart sensors to perform ML inference on every sensor (the eyes of the system) using remote firmware over-the-air updates on every vehicle.
Machine learning, python programming, Tensorflow, Tensorflow lite, Tensorflow lite micro, PyTorch, Keras, computer vision, object detection, image recognition, electronics circuits, embedded systems, MEMS sensors, RF wireless, 5G communications, telematics, FOTA, GitHub, GitLab, ARM Cortex MXX, ARM Cortex AXX, AWS Cloud, AWS IoT, Tableau, Database systems, schematic capture, PCB design and layout, requirements gathering and analysis.