Auto-Tuning parameters of motion cueing algorithms for high performance driving simulator based on Kuka Robocoaster

Area of Science:

• Robotics and Human-Machine Interaction
• Automotive Engineering
• Control Systems

Background:

• Driving simulators are crucial for product testing, with costs varying widely.
• Industrial robots offer a flexible and cost-effective platform for motion simulation.
• Existing motion cueing algorithms (MCAs) are often limited to Cartesian coordinates and struggle with rotational motion artifacts.

Purpose of the Study:

• To integrate and adapt MCAs for a novel industrial robot motion simulator platform.
• To develop and evaluate an auto-tuning method for MCAs on this new platform.
• To reduce false motion cues and improve the fidelity of simulated driving experiences.

Main Methods:

• Implementation of classical and cylindrical coordinate (ClCy) MCAs on an industrial robot platform.
• Development of a motion conversion process to adapt MCAs for the robot's workspace.
• Application of Mean-Variance Mapping Optimization (MVMO) for auto-tuning MCA parameters.

Main Results:

• Demonstrated that MCAs can be effectively applied to the novel robot platform with proposed motion conversion.
• Auto-tuned MCAs successfully exploited the platform's workspace, eliminated false angular velocity cues, and compensated for longitudinal acceleration.
• The MVMO auto-tuning method proved transparent and capable of manipulating simulated quantities based on tuning goals.

Conclusions:

• MCAs can be successfully adapted for industrial robot-based driving simulators.
• Auto-tuning MCAs with MVMO significantly enhances simulation fidelity and reduces unwanted motion artifacts.
• This approach offers a more efficient and effective method for developing high-fidelity driving simulators.