IoT and Robotics tools and technologies
- ROS
- TF lite
- C++
- NuttX
- Raspberry Pi
- PX4
- Gazebo
- Mavlink
- Fast DDS
- Fast Buffers
- Telegraf
- Tasmota
- TimescaleDB
- Packer
- Google Coral
- STM32
- Openthread
- Netdata
- Thingsboard
- Intel OpenVINO
- TF Lite Micro
- Intel Movidius
- AirSim
- Gobot
- FedML
- Emqx
- Preempt RT
- FreeRTOS
analysis
A peek into our IoT implementation practices
- Optimized fog algorithms
Designing fog computations is essential to IoT implementations as knowing what to send to the cloud and what to send to a near-by edge device for analysis could make all the difference in the performance of IoT products.
- Cognitive IoT architecture
To enable and empower our teams to build "intelligent things", we design software around our multi-agent cognitive IoT architecture where ML models and neural networks are considered first class citizens.
- Cryptographic primitives
We wrap IoT data with lightweight cryptographic primitives, such as one-way hash compression functions along with cryptographic protocols enforce security controls and ensure no tampering of data would ever occur in transit.
All-round connectivity with Internet of Everything (IoE)
Internet of Everything (IoE) is a new paradigm that goes beyond the boundaries of the traditional Internet of Things (IoT). In the case of IoT, communication messages get transmitted from a machine to machine (M2M) such as the exchange of sensor data across edge devices.
IoE is different in a way that people and business processes and events from daily life and surrounding environments are also integrated in the communication process forming a mesh of decentralized fog networks. Through this realtime communication, algorithms on edge devices, wearable device and mobiles can convert collected information into actions and facilitate data-based decision-making to provide new capabilities and richer experiences.
The next wave of cognitive AI applications will capitalize on the role of IoE in our daily lives, and introducing 5G in our cities will make these application even more reliable and accurate in their decisions and recommendations.
Getting Robots out of their cages
Adopting a human-centric approach and understanding of user patterns is what guides us when we implement software for AI-powered robots.
Equipping robots with ML models and AI to analyze sensor data, recognize people and objects, and avoid obstacles is our way of enabling robots to take over monotonous and dangerous tasks that are not suited for humans.
