Performance evaluation and real-world testing of a pre-production ADAS

Once primarily focused on vehicle production, car manufacturers have recently evolved into comprehensive engineering firms, responsible not only for vehicle styling and mechanics, but also for software development. An escalating number of electronic components have indeed been integrated into vehicles, particularly in Europe, driven by the commitment of enhancing road safety. This trend has fueled the development of Advanced Driver Assistance Systems (ADAS), a suite of software features aimed at improving comfort and safety by providing both active and passive assistance on-board sensor based. One of the most prevalent systems is Adaptive Cruise Control (ACC), an advanced version of conventional Cruise Control (CC). Unlike traditional CC, which merely maintains a constant speed, ACC incorporates automation to adjust the vehicle's speed autonomously. This adjustment is achieved through the modulation of throttle and brakes, thereby ensuring a safe following distance from preceding vehicles, without necessitating manual intervention from the driver. Toyota seeks to advance this technology one step forward, by enabling speed adaptation based on encountered landmarks such as junctions, curves, and toll gates, requiring the driver only to control the steering wheel. To achieve such a result, an efficient map integration, an accurate camera recognition and a fine tuning are essential. This study have been conducted during the prototype function testing phase. Utilizing a test vehicle equipped with an experimental version of ADAS, an extensive testing campaign was carried on. Collected data have been deeply analysed through an internally developed tool chain, identifying issues and extracting key performance indicators (KPIs). Additionally, competitor vehicles equipped with similar systems underwent testing as part of bench-marking activities, aimed at assessing software quality and exploring solutions offered by other companies. This research facilitated Toyota in identifying critical targets and prioritizing improvements for the enhancement of its map-based Adaptive Cruise Control. Over 20 distinct issues were accurately identified, their root causes determined, and pathways for resolution delineated. Furthermore, in-depth investigations into specific problems enabled the proposal of new logic implementations, unlocking enhanced responsiveness and convenience in the system's behavior.