BEGIN:VCALENDAR VERSION:2.0 PRODID:-//TU/e ME//Events//EN CALSCALE:GREGORIAN METHOD:PUBLISH BEGIN:VTIMEZONE TZID:Europe/Amsterdam BEGIN:STANDARD DTSTART:19701025T030000 RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU TZOFFSETFROM:+0200 TZOFFSETTO:+0100 TZNAME:CET END:STANDARD BEGIN:DAYLIGHT DTSTART:19700329T020000 RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU TZOFFSETFROM:+0100 TZOFFSETTO:+0200 TZNAME:CEST END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:event-11@tuemeche.nl DTSTAMP:20261008T003406Z DTSTART;TZID=Europe/Amsterdam:20260211T133000 DTEND;TZID=Europe/Amsterdam:20260211T150000 SUMMARY:Risk-based Motion Planning in Automated Vehicles DESCRIPTION:Speaker: Leon Tolksdorf\nHost: Nathan van de Wouw\n\nIt is an ticipated that there will be a long period where automated vehicles (AVs) and human road users share the same road network\, with limited communic ation between them. As a result\, AVs often operate with uncertain knowle dge of their surroundings due to limitations in both sensing and estimati on of the traffic scenario's current and future states. These limitations can be caused by factors such as noisy sensor data or lack of informatio n about the intentions of other road users. Accounting for this uncertain ty while designing motion planning and control algorithms for AVs is crit ical\, as an AV may display hazardous behavior if it is overly confident in falsely estimated environment states or\, contrarily\, may act too cau tiously when considering unnecessarily large safety margins in estimated states. Although safety regulations call for taking uncertainty into acco unt\, the right means to integrate uncertainty in motion planning has yet to be agreed upon. A promising way to account for uncertainty is through the concept of risk. Risk accounts for uncertainty in assessing safety b y considering both the likelihood and severity of unsafe events\, such as collisions. Research has shown that human drivers navigate traffic by tr ading off subjectively perceived risk against the potential rewards of re aching their destinations. Additionally\, studies indicate that subjectiv e risk tends to correlate with objective risk measures\, while ethicists argue that risk can be used as a means to facilitate ethicality in road t raffic by distributing risk fairly among road users. These relationships highlight the benefit of incorporating risk\, as a proxy for uncertainty in safety\, in motion planning algorithms.\nIncorporating risk in AV moti on planning presents two main challenges. The first challenge is to devel op a notion of risk that correctly incorporates an AV's uncertainty about the future motion of other road users\, and that can be computed fast (a nd precisely) enough to be used in online (i.e.\, real-time) motion-plann ing algorithms. The second challenge is to develop an appropriate AV moti on planning algorithm that incorporates risk and generates the desired AV behavior. This dissertation addresses both challenges by making five mai n contributions. \nFirst\, we introduce an appropriate risk metric: the e xpected severity of a collision event. This metric\, thus\, is constitute d of two terms\, one accounting for the likelihood of a collision and ano ther term considering the severity of a collision. Additionally\, we inve stigate the AV behavior given optimization-based motion controllers that accommodate our risk metric. We find that a stochastic model predictive c ontrol (SMPC)-based motion planner adequately balances risk with travel e fficiency in the presence of varying uncertainty.\nSecond\, to compute ri sk\, we provide an computationally efficient (i.e.\, real-time computable ) algorithm to estimate risk. Here\, the focus lies on estimating the pro bability of collision (POC)\, as our risk metric reduces to a POC metric when the severity term is considered constant. Based on multi-circular sh ape approximations of the geometries of the AV and other road users\, we present an algorithm that estimates the POC with good precision\, and tha t is guaranteed not to under-approximate it. The algorithm is shown to ex ecute significantly faster than standard Monte Carlo sampling. Further\, we use the proposed algorithm within a SMPC-based motion planning strateg y as a constraint and show that the SMPC generates smooth\, reproducible trajectories while the controller deals with varying levels of uncertaint y.\nThird\, we extend our POC algorithm to risk estimation. Here\, we pro pose a severity estimation method that allows us to consider different co llision constellations\, as literature shows that those are associated wi th different levels of harm to the passengers of vehicles. The resulting risk estimation algorithm leverages the advantages of the POC estimation algorithm\, i.e.\, it is particularly suitable for optimization-based mot ion planning strategies\, while allowing for the flexible integration of severity functions to represent different collision types. \nFourth\, we provide an extension of our risk metric to account for the risk from the individual perspective of each road user in the driving scene.\nGiven a p roposed SMPC-based motion planning strategy\, our AVs behavior can be tun ed such that it minimizes its own risk (egoistic perspective)\, the risk other road users face (altruistic perspective)\, and a balance of both (c ollective perspective). To evaluate safety and travel efficiency (i.e.\, reaching a desired destination in adequate time) for each of these risk m etrics\, we performed simulations on a database containing a wide variety of real-world and artificial traffic scenarios. The results show that a collective perspective balances the overall risk of all road users with t he AV's travel efficiency in a favorable manner\, and constitutes a suita ble approach to facilitate ethicality in road traffic.\nLastly\, we use o ur proposed risk metric as part of the decision-making process on emergen cy maneuvers\, i.e.\, emergency steering and emergency braking\, for an A V designed to evade collisions with late-detected pedestrians. In simulat ions\, we show that this metric outperforms deterministic risk estimation by choosing the appropriate evasive maneuver more frequently. In experim ental testing\, we verified the real-world applicability of this metric i n an AV testbed designed for collision avoidance by having a pedestrian d ummy walk out from a sight obstruction when the AV approaches. We find th at risk is beneficial to deciding on accident avoidance maneuvers.\n\nThe contributions introduced in this thesis show that risk\, and thus uncert ainty\, can be systematically incorporated in AV algorithms for motion pl anning and decision-making. The proposed algorithms are computationally e fficient for real-time optimization-based motion planning\, navigating a fully automated vehicle smoothly through road traffic. Nevertheless\, we also show that our risk metric is beneficial to the avoidance capabilitie s of low-automation safety systems. Furthermore\, we show that risk-based AV behavior generation enables balancing the AV's interests in safety an d travel efficiency for all road users\, and thus\, facilitates ethicalit y.\n\nMore info: https://www.tue.nl/en/news-and-events/news-overview/12-0 2-2026-balancing-safety-and-efficiency-in-automated-vehicles-with-risk-ba sed-planning LOCATION:Atlas 0.710 URL:https://tuemeche.nl/peoplepages/event.php?id=11 CATEGORIES:PhD Defense END:VEVENT END:VCALENDAR