aiSim Features | aiMotive
Core features
Our latest video is showing how our customers can develop their perception and automated driving SW in very fast iteration cycles using our re-simulation solution, which can automatically recreate real-world scenarios in simulation based on the auto-annotation of recorded data. Re-simulation is also unique in allowing customers to recreate any issues arising from real-road testing and explore the capabilities of their SW in a virtual closed-loop environment.
Real-time environment manipulation
Simulating road degradation, lane marking deterioration and other road features
Weather conditions
In the latest version of aiSim, you can simulate sunny, foggy, rainy or snowy conditions and alter them, even making micro-adjustments in real-time, during a simulation, such as combining sunny weather with wet asphalt or poor visibility – all with just a few clicks.
Camera types
Visualizing a broad range of camera types & related ground truth information in aiSim to speed up camera perception development
Synthetic data
Synthetic data generated by aiSim can improve the performance of NN-based perception algorithms in a fast, safe, and cost-efficient way for highway autopilot applications
Sensor configuration
Interactive sensor configuration and physics-based weather simulation to maximize ODD coverage
Sensor calibration
Virtual testing of sensor calibration algorithms
Closed- and open-loop simulation to test and validate AD functions
Test and validate ADAS/AD stacks in diverse traffic scenarios, environment and road conditions and different ODD with fully deterministic and physics-based simulation.
Hardware-in-the-Loop simulation
Verify system functionalities on the target ECU with deterministic, real-time operation mode integrated with the physical interfaces of sensors and communication by saving valuable mileage in test vehicles and reveal corner cases. aiSim has been integrated with a variate of hardware suppliers, including b-plus, dSpace, and National Instruments.
Large-scale scenario generation & Adaptive testing
Adaptive testing enables the dynamic adjustments of test scenarios based on the AD/ADAS stack’s behavior. Explore and locate edge-cases automatically to understand the limits of the AD functions.
Synthetic data generation to address data gaps
Generate synthetic sensor data with plausible scenarios and event sequences, and vary environmental and road conditions. Address the gaps in your data and explore interesting events to ensure the diversity of your datasets.
Multi-sensor simulation for perception testing and validation
Simulate complex sensor setups, including cameras, lidars and radars, using raytracing for ultra-high fidelity visual quality or rasterization for real-time performance. Test your sensor placements and iterate your setup in aiSim to speed up the development and validate your designs.
