Racing team accelerates real-time analysis by 400%

Racing is all about speed, smarts, study, and skill. A big factor that decides how a race will turn out is the pit stop. Each stop should be fast, focused, and efficient. There is no time for errors or miscommunication—racers need to get back on the track quickly, so they don’t lose position.

A high-tech client serving auto racing teams needed to optimize their pit crew’s performance, to get drivers back on track faster and safer. We stepped in to analyze dozens of critical factors—including fuel, tires, weather, crew performance, real-time track conditions and more.

Racing toward the checkered flag

The Ascendion team used AI-powered models and went to work analyzing and processing large volumes of past pit stop data, including historical race data, telemetry, and real-time sensor inputs. We also analyzed factors such as fuel consumption, tire wear and weather conditions.

The key to our solution was our real-time video analysis, using a neural network in Edge devices. We focused on automated continuous learning and tuning. The Ascendion team also implemented a 3D rendering process through multidimensional real-time object detection.

By analyzing the vast amount of data, we were able to generate valuable insights for race strategy, provide realistic scenarios, find areas of improvement and deliver real-time feedback.

Taking a victory lap

Our AI-powered models accelerated real-time analysis by 400% while reducing errors. Our use of AI helped us streamline processes, reduce unnecessary movements, and optimize the coordination and communication among crew members. With improved reliability and minimized errors during pit stops, the need for added stops or penalties was reduced. We were also able to recommend proper training, standardized procedures and safety equipment to minimize the risk of accidents.

Going forward, the racing team can extract valuable insights from race data, assess performance trends, and adjust strategies on the fly to optimize race outcomes. Additionally, the crew is now alerted to potential risks in pit lane activities with real-time alerts and warnings.

Our solution saw the following results:

• 70% effort savings in analyzing the video feeds from the pit crew activities
• Eliminating the human errors in manual video analysis
• 4X time savings in analyzing the video in real-time