Whatever the sector (transport, defense, aeronautics, space, agriculture…), the need for autonomous or semi-autonomous robotization is evolving in parallel with the evolution of technology.
Whether in terms of sensors or computing capacities such as GPUs or FPGAs, uses are exploding.
The aim is either to augment human capabilities, or to dispense with them altogether.
In the field of transport, autonomous vehicles have the potential to radically transform the way we travel, improving the safety, efficiency and accessibility of transport while offering significant economic and environmental benefits.
In space, autonomous systems are crucial to overcoming the unique challenges of space exploration and operations, enabling safer, more efficient and sustainable missions in remote and hostile environments.
Autonomous systems in the space domain are being developed to meet various challenges and requirements specific to space exploration and the exploitation of space resources. Here are some of the main reasons why autonomous systems are essential in the space domain:
In the defence sector, autonomous systems offer significant advantages in terms of safety, efficiency, precision and flexibility. They reduce the risks to soldiers, improve operational capabilities and maintain technological superiority over potential adversaries.
We are also deploying systems that enhance human capabilities (night vision, etc.).
In agriculture, autonomous systems offer significant advantages in terms of efficiency, precision, sustainability and safety. They make it possible to meet current and future challenges in the agricultural sector, while contributing to more sustainable and profitable food production.
In aeronautics, autonomous systems bring significant advantages in terms of safety, efficiency, cost and innovation. They help to meet the industry’s current and future challenges, while opening up new possibilities for flight operations and passenger services.
In industry, autonomous systems offer considerable advantages in terms of productivity, quality, safety, flexibility and efficiency. They enable companies to remain competitive in an increasingly demanding and constantly changing global environment.
Autonomous systems in industry offer many advantages, meeting a variety of needs in terms of productivity, efficiency, safety, quality and flexibility. Here are the main reasons why autonomous systems are increasingly being integrated into industrial environments:
As we saw earlier, depending on the situation and the state of the art of current technology, we are seeking to design semi-autonomous systems (which augment the human being) or autonomous systems (which make it possible to do without the human being altogether).
So what’s the system supposed to do? What is its mission? What are its constraints?
Autonomous systems are distinguished by their ability to perceive, decide, learn and act independently. They combine advanced technologies in perception, artificial intelligence, robotics and communication to offer efficient, safe and adaptable solutions in a variety of fields.
For a system to be autonomous or semi-autonomous, it must be able to carry out a mission in a constrained environment.
The mission may be :
The constraints to be taken into account in system design may be at the level of :
For an autonomous or semi-autonomous system to carry out a mission, it must be able to “see”, understand the environment and situation in which it is evolving, move, transmit information, make a decision, act…
To see and understand the situation in real time whatever the conditions or situations, it is necessary to use several technologies (camera – radar – lidar) and merge the results with powerful onboard multi architectures processors (CPU/GPU/NPU/FPGA/MANY CORE/VPU/ASIC).
Ones need to master the design of the entire vision system chain (optics, image sensor, lighting, embedded HPC electronics (up to 6000 Tops, 100 Gbits/s), software, algorithms, AI, ergonomy, HMI, communications, …).
This cannot be achieved by combining off-the-shelf modules. To achieve performance, it is necessary to design dedicated, integrated systems (See COTS VS CUSTOM). One size fits all never exist.
And even if it’s technologically complicated, we need to be able to design systems that are easy to use.
It's easy to use but bloody complicated to design.
Vincent CARRIER, CEO NEXVISION
Technologies are evolving very rapidly (laser, meta optics, free form optics, processors, image sensor sensitivity, AI models, and, more generally advances in microelectronics (engraving finesse, etc.)…), so we can no longer afford to design products with a 20-year lifespan, because they will be out of date in just a few years.
So, It is necessary to use the latest technologies to avoid becoming obsolete too quickly, while at the same time using technologies that are not yet well mastered takes a great deal of time, effort and investment.
Ones need to challenge himself, move fast, be agile and accept to use the latest technologies coming out of the laboratories. It’s a key success factor.
To make a good service (super simple to use), you need to master the hardware (not only the software).
That’s what we do at Nexvision.
Our teams design complete vision systems, from hardware to software, including advanced image processing and analysis algorithms.
From the outset, as technology enthusiasts and explorers, we have always sought to master the entire chain of vision systems.
Today, we can be proud of our teams.
We don’t hesitate to use the latest technologies straight out of the laboratory to keep our customers one step ahead.
The world in which we operate is highly competitive and changing extremely fast. It’s vital to be on the lookout for the latest technologies, to grasp them and integrate them into architectures as early as possible. And even if it’s often painful because you have to spend many hours understanding, researching and testing, it’s all about being resilient, pugnacious and persevering.
That’s the key to success.