Environment Model & Automated Driving
Partly and fully automated driving takes the driver out of the „sense-plan-act“ equation for selected driving tasks. To accomplish this safely, it is crucial to derive a consistent and dynamic view and interpretation of the vehicle’s surroundings. This is achieved by fusion of sensor, vehicle odometry, and network data into one comprehensive environmental model. Based on the interpretation of this model, path planning strategies are applied and translated into driving commands.
At CES, we are deeply involved in environment model development with our internal partners and external customers.
Based on our expertise in image-processing and radar-signal-processing, we offer core building blocks for Environment Models. Including algorithms for object tracking and SLAM, based on any kind of sensor input. This is complemented by our expertise in trajectory planning and motion control applied in our Automated Driving projects.
Our product and project portfolio for environment modeling & automated driving includes:
- Software Products, proven in the field and continuously improved
- Self Localization & Mapping (SLAM) based on radar, camera or lidar input
- Object tracking
- Object classification (pedestrian classification, two-wheeler-classification, vehicle classification, and obstacle classification)
- Sensor-based ego speed estimation
- Path prediction
- Sensor development kits for fast low-level data access from radar, camera, and lidar sensors
- Cabin and driver monitoring sensors and software
- Commissioning of Automated Vehicles, including
- Trajectory planning & Motion control
- Integration / access to drive-by-wire-system
- Built-up of entire AD-vehicles
For automated driving functions, 360° sensor coverage and central data fusion becomes inevitable. We offer a modular, off-the-shelf prototype solution for your development activities. A set of different hardware and software modules fit the kit for your individual application.