Self-Driving Vehicle System

Machine Learning

LIDAR (Light Detection and Ranging) and cameras are some of the critical hardware sensors supporting the success of self-driving autonomous vehicle systems. These “eyes” of the autonomous vehicle system pull in raw data, which must then be safely and accurately interpreted.

We achieved state-of-the-art results in odometry estimation by applying Machine Learning algorithms for object detection and segmentation (cars, pedestrians, road markup, etc). In order to refine the estimations obtained from algorithms based on LIDAR data, we added fusion with data obtained from GPS/IMU/telemetry and used cameras as an additional input to make it full 3D (6-DOF).

Moreover, we applied generative adversarial network and capsule network models to learn POV-invariant descriptors of local surface features. This helped to safely detect road signs and signals during inclement conditions, such as a rainy night.

Technologies

Sensor fusion (LIDAR, IMU, GPS, Camera)
Odometry Estimation
Object Segmentation
Capsule Network
Python
PyTorch
gRPC
Robot Operating System(ROS)
RANSAC
Object Recognition
Generative Adversarial Network
Performance Optimization
Real Time Systems
C++
Tensorflow Serving
OpenCV

RELATED CASES:

Voice Identification

This system records conversations between the patient and triage personnel and between providers and triage personnel to quickly identify callers and provide for an efficient triage process. If the caller cannot be identified by phone number, the system will use voice fingerprint.

Voice Technology and Chat Bot Machine Learning

Virtual Dispatch Assistant (VDA)

Virtual Dispatch Assistant automatically schedules caregivers for assignment, taking into consideration various conditions and criteria while maximizing the efficiency and utilization.

Voice Technology and Chat Bot Machine Learning

Virtual Triage Assistant (VTA)

We built a Triage Assistant сhatbot that supports the triage nurse in handling patient triage. The chatbot can identify protocols based on patients’ input and conduct all the triage activities. This solution refines itself by reviewing representative data.

Voice Technology and Chat Bot Machine Learning

Drug Identification

One of the challenges caregivers face while visiting patients’ homes is identifying and recording medical/pharmaceutical information in an accurate and efficient way. Proxet builds solutions which rely on image recognition functionality and may allow caregivers to shorten the time of their visit.

Healthcare AI/ML R&D Machine Learning

Medical Emergencies – Chief Complaint Extraction

To reduce unnecessary emergency visits, this platform evaluates patients’ health concerns over the phone and makes recommendations. Once a patient makes a triage call to a nurse to discuss their medical concerns, a chief complaint is recorded and evaluated in the system to provide the triage nurse with appropriate Schmitt-Thompson protocols.

Healthcare AI/ML R&D Machine Learning

Sentiment Analysis for Marketing Campaigns

This project improved the efficiency of marketing campaigns by automatically evaluating user feedback that comes through different channels and automating decisions on further user engagement.

Machine Learning

Advertising Optimization Module

Businesses may find online advertising complex and confusing. One challenge is to maximize results in a cost-effective manner. Based on historical data from advertising campaigns, we developed a response model to simulate different types of custom audience behavior.

Marketing Machine Learning

Image Recommendation Platform

A critical goal of content sites is to keep users engaged for an extended period of time. With this in mind, we have extracted GIF view data from user sessions over the past two years and built a recommendation system that suggests similar GIFs based on users’ areas of interest.

Machine Learning

Facial Recognition of Celebrities Within GIF Graphics

Users frequently search for celebrity names and expect to see all the GIFs associated with particular celebrities. However, not all creators properly tag the GIFs they upload. This means that user searches may not bring up accurate results.

Machine Learning

NLP Solution for Medical Research Databases

Our easy-to-use solution lets medical researchers query data using natural English language either as text or voice and converts it into a structured SQL request.

Healthcare AI/ML R&D Voice Technology and Chat Bot Machine Learning

Complex Flight Planning Tool

We implemented a complex flight planning tool which computes optimal aircraft routes over various geographical objects while taking into account fueling time and fuel consumption, number of aircrafts in a fleet, no-fly zones and other special conditions.

Aerospace Machine Learning

Service Enabling Platform

We developed a modern service enabling platform that solves the communication gap between enterprise brands, 3rd party contractors and consumers. The system connects brands to an army of on-demand local contractors, provides contractors with an intuitive platform to assign and manage jobs and automatically communicates service details to the consumer via mobile devices.

Startup Partnership Machine Learning Ruby

GIFs Search Engine and Performance Optimization Project

The high-level goal of this project is oriented towards improving a search engine to better understand the user’s intent behind the search query and suggest the matching imagery. To achieve this goal, our team has designed, built, validated and trained an API that encompasses several Machine Learning algorithms.

Machine Learning Scala

Media Campaign Optimization Software

Programmatic Marketing software enables marketers to better understand and engage customers across a wide range of media formats, devices, channels and buying modes.

Marketing Machine Learning Java

Leading Marketing Provider

Self-Driving Vehicle System

Technologies

RELATED CASES:

Stand with Ukraine