xCog: QA for AI, Robotics, and Autonomy
Quality Assurance for AI, Robotics, and Autonomy
Currently, automated testing for cognitive systems is a relatively new process that requires further exploration. As technology evolves at a rapid pace and gets “smarter”, so do testing methods. Just a few years ago, the leap from a facial recognition bot to machine-based surgery would have been considered science fiction. However, these days there’s more science than there’s fiction as the ideas have become reality.
The need for AI, Robots, and Autonomy is growing by the day. While the human population grows, so does the need for technology to accommodate in ways that allow for a higher quality of life. Machines already play a vital role in the human world, with traffic lights, fraud detection, chatbots, stock trading, agricultural pest control, and the list goes on.
With all of these new avenues and technological advances one question rises to the top, is it safe? Will AI understand human error versus fraud? Can a robot perform surgery and make complex decisions? How can an autonomous vehicle recognize a person crossing the road outside of their programming?
QA Consultants wants to answer these questions and more need by offering a solution. Where to begin? The following are very real issues that are a good place to start a conversation.
Current autonomous robot vehicles quality challenges
Limited knowledge
Lack of expertise to properly test autonomous robot vehicles
Regulatory compliance
Regulatory compliance and government integrations or approval
Long time to market
Time spent testing and validating hazard scenarios is lengthy
The high cost of physical tests
It is not feasible to physically reproduce several test scenarios
Risk for the robot buyers
No formal guarantee that the vehicle can avoid collisions with objects or people
Limited test coverage
Uncertain about the robotic vehicle’s behaviour in different situations
Late validation in SDLC
Validation and testing are done in the late stages of the robot vehicle development
High complexity
High degrees of complex combinations to be tested to ensure high quality
The Solution
At QA Consultants, we’ve listened to our clients explain their pain points and we’ve created solutions.
- A proprietary model-based solution that automates the generation of automated test cases for cognitive systems
- Integration with industry-standard simulation software
- Automated execution of test cases and result analysis
Accelerating time-to-market with dramatic cost reduction
Benefits & advantages of the xCOG solution
-
Lower cost due to simulation
Any test scenarios can be performed in a simulated environment
-
Fast time to market
Hundred of thousands of different test scenarios can be executed quickly
-
Quality & safety assurance
Potential issues can be identified and fixed before the robotic vehicle goes to production
-
Strategy for a real-world UAT
Focus on scenarios identified as higher risks to be reproduced in real-word with physical tests
-
Lower risk for the robot buyer
With a formal quality assurance process, robot buyers can get confidence that it is ready to be interacting in a real environment with people without hazard risks
-
High testing coverage
The model-based testing approach allows that hundred of thousands of test scenarios are automatically generated and executed
-
Early validation in SDLC
The software that controls the robotic vehicle can be tested in the early stages of development, due to the automated approach and easy validation
Accelerating the automated testing for cognitive systems
-
Faster and economical
The setup of testing is faster and more economical, once it doesn’t require physical equipment and environment
-
Model-based testing
Generation and prioritization of test cases are automatically done, using a model-based approach
-
High scalable
The test is performed in a simulated environment in the cloud
-
Full automation
Automated tests are automatically generated by the framework
-
Velocity and permutation
Safety through velocity and permutation: A solution that has automated the creation of automated tests
-
Safer test execution
No risk to human life or equipment to validating thousands of scenarios that are needed
Market uses
End-to-end Quality
Government & regulators
Government, regulation entities and even insurance companies might be interested in making sure robot vehicles are safe enough when in contact with people
Robot manufacturers
Companies that build a physical robot vehicle. These companies usually are responsible for the entire robot vehicle and the first point-of-contact for buyers or users
Robotic systems owners
Companies that buy robot vehicles and have them running in their environment. Robot Buyers include airports, malls, government agencies, warehouses, hospitals, etc.
Funders & investors
Organizations that provide funds and grants for the autonomous robot vehicles ecosystem to invest in technology and solutions that address different needs
AI developers
Software (AI) builders that control and provide autonomy to robotics vehicles.
How can we get your business to the next level?
Let’s talk. Fill out the form and let us know what you’re interested in doing.