DAVI Progress & Ambitions Dutch Automated Vehicle Initiative www.davi.connekt.nl R. Happee r.happee@tudelft.nl Dutch Automated Vehicle Initiative – 2014 1 Dutch Automated Vehicle Initiative To investigate, improve and demonstrate automated driving on public roads • • • • • Assess & improve technology Study human behaviour Prove safety Pursue legalisation Create public awareness Dutch Automated Vehicle Initiative – 2014 2 DAVI in time • • • • Inception Official Launch Innovatie-estafette Experiments with non-expert drivers Demo with non-expert drivers Dutch Automated Vehicle Initiative – 2014 March 2013 Nov 2013 2015 2016 3 key projects & partners DAVI open initiative TU Delft, RDW, TNO, Connekt, SWOV, UT, Fontys, Hanze, HAN, RUG, Toyota Motor Europe, NXP, Imtech, VisLab (IT), SKF, Technolution, Almende, V-tron, Trinité Mapscape, DLR, TRL, ITS-Leeds Human Factors of Automated Driving (HFAuto) Marie Curie FP7 TU Delft, TU München, Univ of Southampton, Univ of Twente, Chalmers Univ of Technology, IFSTTAR (France), VTI (Sweden), Volvo Truck, Volvo Car, BMW, Jaguar, Toyota Motor Europe, Continental, TNO, SWOV Truck Merging Support HTSM TU Delft, TU/e, DAF, SKF, TNO, NXP From Individual Automated Vehicles to TU Delft, Toyota Motor Europe, TNO, NXP, Imtech Traffic Cooperative Traffic Management & Infra, RDW, Connekt, SWOV, Technolution, Almende, (IAVTRM) STW-OTP V-tron, Trinité Automation, VisLab Foodvalley Dutch Automated Vehicle Initiative – 2014 Province of Gelderland 4 Hands free • Cruise Control with automated steering • Highway in 2020 • Capable driver Autonomous • No steer & pedals • User only has to select the destination Mixed with normal traffic Dutch Automated Vehicle Initiative – 2014 5 Vehicles • 3 Priuses TNO / DITCM • CACC • Automated steering • Prius Fontys • Automated steering • 2 Priuses TU Delft • Towards 360 deg sensing Dutch Automated Vehicle Initiative – 2014 6 Ambitions • Vehicle automation • Robust environment perception • Safe control strategies • Link automation to traffic management • Human factors • Transitions of control • Acceptance, trust, use & misuse • Interaction with other road users • Benefit prediction • Legalisation • Public awareness & acceptance Dutch Automated Vehicle Initiative – 2014 7 Environment Perception • Cameras, Radar, Ultrasound, Lidar • • • • Cars, trucks, pedestrians, cyclists Road markings Road Signs Traffic lights • Challenges • Reliable detection in various weather conditions • Intent recognition & motion prediction • Integration environment sensing with communication and detailed maps Dutch Automated Vehicle Initiative – 2014 8 Custom of the shelf sensors ARS 309-2 HS radars (Continental) HDR vision systems (Melexis) High accuracy Digital maps & eHorizon systems (Mapscape – Maccom - Electrobit) Dutch Automated Vehicle Initiative – 2014 Main Processing Unit PXIe-1062Q RTOS (NI) 9 Graceful degradation of CACC πref,4 5 π£5 πref,3 4 3 π5 π£4 2 πref,2 π£3 3 1 π4 πref,1 2 π3 π£2 1 π2 ACC: no V2V communication CACC: full V2V communication Graceful degradation in case of packet loss: dCACC CACC ACC 10 dCACC π£1 Automated lane changing & merging • Shown in movies carmakers • Proof of safety not found • DAVI simulation studies • Optimize gap & timing1 • Optimize collective benefits1 • Trajectory control with non-linear vehicle model2 • Next • Sensors fusion & inaccuracy • Role of the driver? 1. 2. Wang M, Hoogendoorn SP, Daamen W, van Arem B, Happee R. Novel predictive approach for unified lanechanging and car-following control. Submitted ISTT21 Gottardis P, Manazza SS. Automated Controlled Vehicle Based on Non-Linear Model Predictive Control Connected to a Safety Path Planner with Online Collision Risk Estimation MSc thesis TUDelft / Polimi 3 Oct 2014 Dutch Automated Vehicle Initiative – 2014 11 Human Factors • Literature survey1 • Extensive simulator studies Highly Automated Driving • Workload affects HAD much larger than ACC • Only expert drivers tested with HAD on public roads (3 studies) • Challenges • Monitor & Manage driver awareness • Develop safe transition procedures & interfaces • Investigate human interaction with HAD on public roads 300 Secondary Task Performance (%) 200 100 0 1. de Winter JCF, Happee R, Martens MA, Stanton NA. (2014). Effects of ACC and highly automated driving on workload and situation awareness: A review of the empirical evidence. TRPF: Traffic Psychology and Behaviour. Dutch Automated Vehicle Initiative – 2014 Control ACC HAD 12 Opinions on automated driving 1 • Diverse / extreme responses • 22% unwilling to pay for fully automated driving • 5% willing to pay more than $ 30,000 • 69% estimated fully automated driving to reach 50% market share before 2050. • concerned about • software hacking/misuse • legal issues and safety 1. Kyriakidis M, Happee R, de Winter JCF. Public opinion on automated driving: Results of an international questionnaire among 5,000 respondents. Submitted. Dutch Automated Vehicle Initiative – 2014 13 Automated Transport Foodvalley (Gelderland) • “De eerste zelfsturende auto’s van Nederland moeten gaan rijden in Ede en Wageningen” (Gelderlander 1 Oct 14). • Planstudie, suitable routes, sensors & control • Innovation • From closed track to public roads • Safe interaction with pedestrians, cyclists, cars • On demand public transport • Safety by • Low speed • Conservative control strategies • Control room monitoring Dutch Automated Vehicle Initiative – 2014 14 Milestones testing automation on public roads • Technical safety assessment • Sensing reliability • Safe operating conditions • Automation linked to traffic management • Behaviour non-expert drivers investigated • Safe transitions of control • Quantify how drivers will use automation • Prove safety taking into account the human interaction • Benefits predicted with traffic flow models • Legalisation Dutch Automated Vehicle Initiative – 2014 15 Questions Dutch Automated Vehicle Initiative – 2014 16 DAVI Partnership Charter • DAVI aims to • • • • • • • • • implement automated vehicles, obtain permission for testing on public roads, link vehicle automation to traffic management, assess and improve automation technology, study human behaviour with automation, prove safety, optimise transport efficiency, create public awareness, and pursue legalisation of automated driving. • DAVI pursues these objectives through complimentary research projects and demonstration activities with varying partnerships and national and international funding sources. Dutch Automated Vehicle Initiative – 2014 17 Vacancies • Vacancy 1 (PhD Candidate) Human interaction with automation • Human machine interface, driver state monitor • Investigate the human interaction with automation • Vacancy 2 (PhD Candidate) Benefits & risks of automated driving • Develop computer models capturing mixed traffic • Predict & optimise benefits • Vacancy 3 (Post Doc) Automation & safety assessment • Automation algorithms, merging, lane changing and overtaking • Safety assessment methods • Vacancy 4 (Engineer) Environment sensing • Systems architecture • Sensors & fusion providing robust 360 degree environment perception Dutch Automated Vehicle Initiative – 2014 18