ADAS & Autonomous vehicles

SHERPA Engineering has been involved since the early 2000’s in automotive ground support activities.

With the introduction of ADAS (Advanced Driving Assistance Systems), SHERPA Engineering has naturally taken position in the modelling, simulation and control of driving assistance systems and autonomous vehicles: cars, buses, shuttles, etc.

To this end, we have widen our activities to perception and localisation, which require specific skills such as the processing of Lidar sensor data, learning by Deep-Learning and reasoning on uncertain data using Bayesian networks.

In order to provide complete solutions, especially for real time applications, we have also enlarged our offer with the realisation of electronic systems.

A HIGH VALUE-ADDED OFFER TO MEET TODAY’S CHALLENGES

ASSETS TO MAKE A DIFFERENCE

«MADE IN SHERPA» SOFTWARE MODULES

Over the years we have developed our own software modules. When combined, they allow us to offer complete demonstrators such as AD System.

This allows us and our clients to complete projects much faster for two main reasons:

• The customer focus is a software module or sensor that can only be implemented in a complete system. Our modules cover most ADAS or AD requirements and will allow the project to focus on the system of customer interest.
• The customer’s focus is a demonstrator type project of a global function in a given environment, such as logistics, valet parking. The content of the project will then be focused on the integration in a given vehicle and the realisation of the use cases requested by the customer.

Example of available modules

• Object detection and recognition
• Multi-sensor data fusion
• Lidar-based localisation in 2.5 D and 3D versions
• Planning and control for peri-urban environments
• Planning and control for valet parking application

PERCEPTION KITS

Perception kits are systems for measuring perception data while driving.
They are designed to be simply placed on the roof of a vehicle and are therefore not very intrusive.

The available sensors are defined for the customer’s application.

Configuration examples are:

• 1 lidar 360, 2 front cameras, 1 GNSS sensor, 1 IMU
• 1 lidar 360, 9 cameras, 1 GNSS sensor, 1 IMU

The data is synchronised and time-stamped for further analysis. Real-time algorithms can also be integrated.
The middleware used can be ROS or RTMaps.

For further information : read the whitepaper
Conception of a high-level perception and localisation system for autonomous driving

ECOSYSTEM & PARTNERS

SHERPA Engineering has entered the ADAS and Autonomous Vehicle ecosystem through 3 main axes:

• Education-Research

  We are founders and contributors to the autonomous vehicle chairs at INSA Rouen and the Pascal Institute in Clermont-Ferrand.

• Scientific Council

  SHERPA Engineering has set up a Scientific Council headed by Jean-Marc Blosseville, Professor Emeritus. This committee allows us to decide on our own research and development areas, and to offer our clients high-level interventions in the projects we carry out for them.

• Joint laboratory

  Together with Logiroad.ai, a company specialising in infrastructure analysis, and the Pascal Institute, we have founded a joint laboratory in the field of perception.

• Aladin consortium

  We are a member of the NextMove Aladin consortium.
  This group of companies has the will to propose common offers to our customers. It allows us to carry out projects by integrating various skills such as connectivity, mechanical integration, operating safety and many others.

• Partnership with INRAE

  SHERPA Engineering and INRAE have built a partnership in the field of agricultural robotics, in the framework of the Agrotechnopole-INRAE. SHERPA Engineering is particularly involved in carrying out the safety tests resulting from the ARPA project.