Change search
Link to record
Permanent link

Direct link
Publications (5 of 5) Show all publications
Söderman, M., Clasen, R., Bergström, G. & Collings, W. (2022). Development of self-driving and control room functions and of external HMI for automated delivery vehicles.
Open this publication in new window or tab >>Development of self-driving and control room functions and of external HMI for automated delivery vehicles
2022 (English)Report (Other academic)
Abstract [en]

Road users may need additional information to a vehicle’s speed and position in lane to understand the behavior and intentions of self-driving vehicles, such as external Human Machine Interface (eHMI), i.e. visual signals (lights) indicating the vehicle's status, behavior and intentions, especially in environments where self-driving vehicles are expected to drive at low speeds, for example, in urban environments. The study in this report developed and implemented self-driving functions and control room functions in an automated delivery vehicle (ADV), as well as developed an eHMI concept to communicate the vehicle’s states, intentions, and behaviour to the surroundings. The software stack and the development of the main features of the self-driving driving capabilities including the lateral controller are described in this report. Further, modules such as the Representational State Transfer (REST) for communication, the remote control of the ADV and the eHMI communication interface with the vehicle signals are presented. A lesson learned from the study is that further refinement in repeatability of the initial conditions of the system is essential. Most of the individual parts of the chain from a command to the ADV were created via the user interface in the Autonomous Transport Management System (ATMS). However, the whole chain was hard to achieve, and the need of frequent testing and integration was evident and faults in the chain could result in cumbersome and long procedures to restart the integration test. The study also revealed issues with the stability of the entire system. Several eHMI concepts have been developed in industry and in research. However, up to this date there are no standards or established frameworks for the design of eHMI. Nevertheless, guidelines and recommendations have been proposed in different studies, for example that eHMI should be consistent with existing eHMI, address road users in general and not tell or instruct other road users what to do. The eHMI-concept developed in this study conveyed the following messages: Automation mode, Acceleration, Deceleration and Delivery mode. A model for an eHMI-strategy is also proposed. The eHMI prototype on the ADV in this study was composed of LED lights with multiple color options, an ECU with CAN hardware and software that controlled the eHMI. The initial idea was to use vehicle data from the CAN bus, such as speed and steering angle and the control algorithm worked technically, but the eHMI for acceleration and deceleration were activated/inactivated too fast (within 1-2 seconds) for an observer to perceive and grasp the meaning of the eHMI. A lesson learned was that the activation/inactivation of the eHMI should, therefore, be executed by the computer that manages the autonomous driving functions in the ADV.

Abstract [sv]

Trafikanter och fotgängare kan behöva mer information utöver fordonets hastighet och position i körfältet för att förstå ett självkörande fordons körbeteende och avsikter, till exempel olika former av externt Human Machine Interface (eHMI), d.v.s. visuella signaler (lampor) som visar fordonets status, beteende och intentioner, framför allt i miljöer där självkörande fordon förväntas köra i låga hastigheter, till exempel i stadsmiljöer. I iteration 1 hade fordonet inga självkörande funktioner. Studien i denna rapport fokuserade på att utveckla och implementera självkörande funktioner för mindre leveransfordon (eng. Automated delivery vehicle, ADV), samt att utveckla kontrollrumsfunktioner. I studien utvecklades även ett koncept för eHMI. Flera eHMI-koncept har utvecklats inom industri och forskning, men det finns inga standarder eller regler för eHMI. Det finns dock riktlinjer och rekommendationer från olika studier om eHMI, vilka har legat till grund för utvecklingen av eHMI-konceptet i denna studie, och som indikerar följande lägen: Automatiseringsläge, Acceleration, Deceleration och Leveransläge. En modell för en eHMI-strategi föreslås också. Den tekniska utvecklingen i studien omfattande bl.a. programvarustacken och de automatiserade körfunktionerna, inklusive sidokontrollen, Representational State Transfer (REST) för kommunikation med eHMI-signalerna och med fordonssignalerna. De flesta av de enskilda delarna av kedjan från ett kommando till ADV:n skapades via ett gränssnitt i det autonoma transporthanteringssystemet (eng. Autonomous Traffic Management System, ATMS). Hela kedjan var dock svår att uppnå, och vikten av testning och integration blev tydlig. Fel i kedjan kunde bl.a. resultera i besvärliga och långa procedurer för att starta om integrationstestet. Studien visade också att stabilitetsproblemen i systemet. eHMI-prototypen på ADV:n bestod av LED-lampor med flera färgalternativ, en ECU med CAN-hårdvara och programvara som styrde eHMI. Till en början användes fordonsdata från CAN-bussen (hastighet och styrvinkel) som aktiverade de olika eHMI-signalerna. Kontrollalgoritmen fungerade tekniskt, men studien visade att eHMI för acceleration och deceleration aktiverades/inaktiverades för snabbt (inom 1-2 sekunder) för hinna uppfatta och förstå innebörden. En slutsats var att aktivering/inaktivering av eHMI i stället bör utföras av samma dator som hanterar de autonoma körfunktionerna.

Publisher
p. 24
Series
Trafikverkets forskningsportföljer
Series
RISE Report ; 2022:132
Keywords
Autonomous delivery vehicles, Self-driving functions, Control room functions, external HMI development
National Category
Transport Systems and Logistics
Research subject
FOI-portföljer, Strategiska initiativ
Identifiers
urn:nbn:se:trafikverket:diva-16257 (URN)978-91-89757-21-9 (ISBN)
Projects
GLAD - Godsleverans under den sista milen med självkörande fordon
Funder
Swedish Transport Administration, TRV 2020/26017
Available from: 2024-08-19 Created: 2024-08-19 Last updated: 2025-09-04
Söderman, M. (2022). GLAD Godsleverans under den sista milen med självkörande fordon - sammanfattning av GLAD-projektet.
Open this publication in new window or tab >>GLAD Godsleverans under den sista milen med självkörande fordon - sammanfattning av GLAD-projektet
2022 (Swedish)Report (Other academic)
Alternative title[en]
Goods deliveries under the last mile with automated delivery vehicles – a summary
Abstract [sv]

Detta är en sammanfattning av studierna som gjordes i GLAD-projektet (GLAD: Goods deliveries under the last mile with automated delivery vehicles). GLAD var ett forsknings- och utvecklingsprojekt som hade det övergripande syftet att generera ökad kunskap om användares behov, samt om de tekniska, affärsmässiga och policyrelaterade utmaningar som automatiska leveransfordon (eng. Automated delivery vehicles, ADV) kan innebära. GLAD-projektet genomfördes under juni 2020 och september 2022 och koordinerades av RISE Research Institutes of Sweden. Projektet delfinansierades av Trafikverket (ref. no. TRV 2020/26017). Parterna i GLAD-projektet var RISE Research Institutes of Sweden, Aptiv AB, Combitech AB, Clean Motion AB och Högskolan i Halmstad.

Abstract [en]

This is a summary of the studies that were carried out in the GLAD project (GLAD: Goods deliveries under the last mile with automated delivery vehicles). GLAD was a research and development project with the overall aim to gain knowledge about user needs as well as about the technical, business and policy related challenges with automated delivery vehicles (ADV). The GLAD project was conducted during June 2020 and September 2022 and was coordinated by RISE Research Institutes of Sweden. The project was partly financed by the Swedish Transport Administration (ref. no. TRV 2020/26017). The partners in the GLAD project were RISE Research Institutes of Sweden, Aptiv AB, Combitech AB, Clean Motion AB and Halmstad university.

Publisher
p. 22
Series
Trafikverkets forskningsportföljer
Series
RISE Report ; 2022:135
Keywords
Fordon, Godstrafik, Informationsteknik, Autonoma fordon
National Category
Transport Systems and Logistics
Research subject
FOI-portföljer, Strategiska initiativ
Identifiers
urn:nbn:se:trafikverket:diva-16253 (URN)978-91-89757-24-0 (ISBN)
Projects
GLAD - Godsleverans under den sista milen med självkörande fordon
Funder
Swedish Transport Administration, TRV 2020/26017
Available from: 2024-08-19 Created: 2024-08-19 Last updated: 2025-09-04
Söderman, M., Clasen, R., Bergström, G. & Collings, W. (2022). People’s understanding of external HMI and their experiences of interacting with an automated delivery vehicle in a terminal context.
Open this publication in new window or tab >>People’s understanding of external HMI and their experiences of interacting with an automated delivery vehicle in a terminal context
2022 (English)Report (Other academic)
Abstract [en]

This study investigated how the participants in a scenario of a terminal for goods handling perceived to interact with an automated delivery vehicle (ADV) for loading/unloading and how they understood the external Human Machine Interfaces (eHMI), i.e., visual signals on the ADV that communicate the ADV’s behaviour, mode and intentions. The objectives were (i) to test and validate the self-driving functions; (ii) to test and validate the control room functionalities and (iii) to evaluate how the participants understood the eHMI and (iv) to evaluate how they experienced to interact with the ADV in specific situations. The eHMI communicated four messages; Acceleration (from stand still), Deceleration (to stand still), Unplanned stop and Delivery mode. The participants were introduced to the scenario and instructed to act the role of being newly employed at a terminal for loading/unloading goods. There were two situations they were asked to handle: (i) the ADV had stopped (unplanned) due to an obstacle in front which had to be removed for the ADV to drive on, and (ii) to load/unload the ADV when it had stopped at a designated place for loading/unloading. The participants marked on a 5-point scale how easy/difficult it was to understand the different eHMI (eHMI communicated Acceleration from standstill, Deceleration to standstill, Unplanned stop and Load/Unload mode) and how safe/unsafe they felt to approach and interact with the ADV. The same procedures were repeated three times. The results showed that the participants thought it was easy to understand the different eHMI on the ADV, specifically the types of eHMI that are on vehicles today, such as hazard lights and turning indicators. The results also revealed that the context, i.e. terminal scenario, the situations, and the work tasks, was a contributing factor to their understanding of the eHMI. The participants’ previous and gained experiences also contributed to their understanding of the eHMI. The participants thought it was safe to approach the ADV, for example to remove the obstacle in front of the ADV, much because they assumed that such close interactions with the ADVs could happen often and, therefore, they assumed it was safe to interact closely to the ADV. The size of the ADV (smaller than a regular car) was also mentioned as a contributing factor. The self-driving functions in the ADV were integrated in the ADV’s system architecture. A challenge was to obtain stability in the system with repeated driving cycles. The Autonomous Transport Management System (ATMS) was put in a cloud service to enable remote testing, and to facilitate repeated integration tests as well as test-cycles with the ADV. The information in the messages sent to the ADV included, for example, the coordinates for the route and the control signals for the eHMI. In addition, a function was implemented to reset the ATMS easily when it entered a faulty state. The control of the LED lights for the eHMI was managed by the main Vehicle Control Unit (VCU) which provided more accurate output from the eHMI compared to using the vehicle data.

Abstract [sv]

Denna studie utgick från ett scenario där ett självkörande leveransfordon (eng. Automated delivery vehicle, ADV) anländer för omlastning i en godsterminal. Syftena med studien var att; i) testa och validera de självkörande funktionerna i ADV:n, ii) testa och validera de s.k. kontrollrumsfunktionerna och iii) att utvärdera deltagarnas (i rollen som personal på terminalen) förståelse av eHMI, d.v.s. visuella signaler på ADV:n som kommunicerar dess status, körbeteende och intentioner, och (iv) utvärdera deltagarnas upplevelser av att interagera med ADV:n i två situationer: att ta bort att föremål framför ADV:n och att last/lossa gods från ADV:n. Deltagarna fick agera nyanställda på en terminal för lastning/lossning av gods och där de skulle hantera två situationer: (i) att plocka bort ett föremål framför ADV:n som hindrade den att köra vidare och (ii) att lasta/lossa gods från ADV:n. Deltagarna markerade på en 5-gradig skala hur lätt/svårt det var att förstå de olika eHMI på fordonet (som kommunicerade Acceleration från stillastående, Inbromsning till stillastående, Oplanerat stopp och Leveransläge) och hur säkert/osäkert det kändes att dels plocka hindret framför ADV:n, dels att lasta/lossa från ADV:n. Varje deltagare upprepade resp. situation tre gånger. Resultaten visade att deltagarna generellt förstod de olika eHMI på ADV:n, speciellt den typ av eHMI som finns på fordon idag, t.ex. varningsblinkers och blinkers. Det visade sig också att kontexten, d.v.s. terminalscenariot, situationerna och arbetsuppgifterna, var viktig för deltagarna för att förstå innebörden av eHMI. Även deltagarnas tidigare erfarenheter, samt den erfarenhet och kunskap de fick under studien bidrog till att förstå de olika eHMI. Deltagarna tyckte att det kändes säkert att närma sig ADV:n för att ta bort hindret. De antog denna typ av situation kunde vara vanlig på en terminal med ADV:er och antog därför också att det var säkert att interagera med ADV:n. Storleken på ADV:n (mindre än en bil) var också en bidragande faktor till att det kändes säkert. De självkörande funktionerna i ADV:n och rutten var integrerad i ADV:ns systemarkitektur. En utmaning var att få stabilitet i systemet med upprepade körcykler. Det autonoma transporthanteringssystemet (eng. Autonomous Transport Management System, ATMS) placerades därför i en molntjänst för att kunna fjärrtestas och för att kunna upprepa integrationstester och testcykler. Informationen i meddelandena som skickades till ADV:n från kontrollrumsfunktionerna inkluderade till exempel koordinaterna för rutten och styrsignalerna för eHMI. Även en funktion för att återställa systemet vid felmeddelanden implementerades. Styrningen av LED-lamporna i eHMI hanterades av fordonets styrenhet (eng, Vehicle Control Unit, VCU) vilket gav en noggrannare beräkning av utgångssignalerna till eHMI jämfört med att använda fordonsdata.

Publisher
p. 28
Series
Trafikverkets forskningsportföljer
Series
RISE Report ; 2022:133
Keywords
Autonomous delivery vehicle, understanding of external HMI, Human – ADV interactions
National Category
Transport Systems and Logistics
Research subject
FOI-portföljer, Strategiska initiativ
Identifiers
urn:nbn:se:trafikverket:diva-16258 (URN)978-91-89757-22-6 (ISBN)
Projects
GLAD - Godsleverans under den sista milen med självkörande fordon
Funder
Swedish Transport Administration, TRV 2020/26017
Available from: 2024-08-19 Created: 2024-08-19 Last updated: 2025-09-04
Söderman, M. (2022). Typical and critical traffic situations with small electric delivery vehicles – indications for future Automated Delivery Vehicles.
Open this publication in new window or tab >>Typical and critical traffic situations with small electric delivery vehicles – indications for future Automated Delivery Vehicles
2022 (English)Report (Other academic)
Abstract [en]

This study investigated what typical traffic situations drivers of small manual delivery vehicles (MDV) are facing during their daily routes and how they handle these, sometimes critical, traffic situations. The purpose was to get an understanding of what challenges future automated delivery vehicles (ADV) may encounter and need to manage. Nine drivers of MDVs at one of Postnord’s terminals in Gothenburg, Sweden, were interviewed about their daily working tasks, their experiences of typical and critical situations and how they handle these situations. The interviews showed that many potentially critical situations were related the MDV’s relative slow speed (max 45 km/h). They could not always keep the same speed as other vehicles, which resulted in other vehicles driving closely behind the MDV and overtaking the MDV in narrow and busy roads. The interviews also revealed that the drivers often need to remove obstacles. Since an ADV cannot solve these kinds of problems like human drivers do the ADVs’ Operational Design Domain (ODD) may need to be adapted to the ADV’s capacity, e.g. being free of obstacles. The letters and packages are delivered to the addressees by the drivers. With ADVs, these “hand-over” operations need to be either taken care of by someone at the addressees or be replaced by a delivery system that does not involve the hand-over to the addressees. Another matter is that some general traffic rules are often vaguely formulated (“… adapt the speed to the bicycles…”, “…adjust the speed so there is no danger…”, “…to… in time…”) and leave much to the drivers to interpret their meanings and to act accordingly. How ADVs should comply with this kind of traffic rules could be a challenge. The drivers’ gained experiences seemed to be key to handle unforeseen events and to solve problems as they occur, for example through compensating behaviour, such as position in lane, acceleration/deceleration, steering manoeuvres etc. A “dynamic learning function” could be an important feature to implement in a future ADV-system. Overall, the interviews showed that the drivers are handling complex traffic situations and environments and that they need to manage many practical tasks to deliver the letters and packages to the addressees. Without human drivers a delivery system with ADVs would require a systems perspective throughout the whole logistics chain.

Abstract [sv]

I denna studie undersöktes vilka trafiksituationer som förare av små leveransfordon (eng. Manual delivery vehicles, MDV) råkar ut för och hur de hanterar dessa, ibland kritiska, trafiksituationer i sitt dagliga arbete. Syftet var att få en bild av vilka situationer som framtida självkörande leveransfordon (eng. Automated delivery vehicle, ADV) kan komma att råka ut (förutsatt att de kommer köra i liknande miljöer). Nio förare på en av Postnords terminaler i Göteborg intervjuades om sina arbetsuppgifter, erfarenheter av typiska och kritiska situationer och hur de hanterar dessa situationer. Intervjuerna visade att många potentiellt kritiska situationer orsakades av fordonets låga hastighet (max 45 km/h) jämfört med andra fordon, t.ex. fordon kör mycket nära bakom dem och vårdslösa omkörningar sker på smala och trafikerade platser. Intervjuerna visade också att förarna ofta måste ta bort hinder för att komma fram. Eftersom en ADV inte kan lösa dessa typer av problem som förare gör idag kan det innebära att ADV:ns Operational Design Domain (ODD) behöver anpassas. De brev och paket som körs ut levereras till adressaterna av förarna. Med ADV:er måste detta antingen tas om hand av någon person hos adressaterna eller ersättas av ett annat system som inte innebär att en person tar breven och paketen den allra sista biten till adressaterna. Studien visade även att förarnas erfarenheter har stor betydelse för att köra säkert och effektivt och att de löser problem allteftersom de uppstår genom kompenserande körbeteende, t.ex. position i körfält, acceleration/retardation, styrmanövrar etc. En "dynamisk inlärningsfunktion" kan därför vara en viktig funktion att utveckla i ett framtida ADV-system. Förarna nämnde även att vissa trafikregler är vagt formulerade (”... anpassa hastigheten så att det inte uppstår fara...", "... anpassa hastigheten till cykeltrafiken...", "... att i god tid...") vilket innebär att de måste tolka innebörden och agera därefter beroende på rådande trafiksituation. Hur framtida ADV:er ska förhålla sig till denna typ av generella trafikregler kan bli en utmaning. Sammantaget visade intervjuerna att förarna hanterar komplexa trafiksituationer utifrån sina kunskaper och erfarenheter, samt att de, förutom att köra fordonet, måste hantera en mängd praktiska saker för att leverera brev och paket ända fram till adressaten. Utan förare måste ett leveranser med ADV:er genomlysas av en systemsyn på hela logistikkedjan.

Publisher
p. 12
Series
Trafikverkets forskningsportföljer
Series
RISE Report ; 2022:130
Keywords
Electric delivery vehicles, Driver actions, Automated delivery vehicles, Traffic situations, Operational design domain
National Category
Transport Systems and Logistics
Research subject
FOI-portföljer, Strategiska initiativ
Identifiers
urn:nbn:se:trafikverket:diva-16254 (URN)978-91-89757-19-6 (ISBN)
Projects
GLAD - Godsleverans under den sista milen med självkörande fordon
Funder
Swedish Transport Administration, TRV 2020/26017
Available from: 2024-08-19 Created: 2024-08-19 Last updated: 2025-09-04
Söderman, M., Andersson, J. & Habibovic, A. (2022). Use cases and high-level requirements for safe interactions between automated delivery vehicles and human operators in a terminal.
Open this publication in new window or tab >>Use cases and high-level requirements for safe interactions between automated delivery vehicles and human operators in a terminal
2022 (English)Report (Other academic)
Abstract [en]

Small electric Autonomous Delivery Vehicles (ADV) can play an important role in future logistic chains under the last mile deliveries. In terminals where ADV are loaded with goods it is important that the interactions between the ADVS and the goods handling personnel is safe. Two workshops with developers of self-driving vehicles, researchers in the area of human-machine interaction and goods handling personnel form Postnord were conducted to identify challenges, needs and requirements regarding the design of ADV and the terminals för ADV. Due to COVID19, the workshops were carried out online and a video was shown to the participants demonstrating an ADV operating in a location representing a terminal. The two main objectives for this study were to gain understanding of the interactions between the ADV and human operators in the terminal and to identify high-level functional requirements for safe and efficient deployment of ADVs in terminals. The identified use cases related to (i) the ADV’s operations in the terminal, from entering to leaving the terminal and (ii) use cases where human operators interacted with the ADV, e.g. for loading/unloading goods. For each use case a high-level functional requirement was formulated. Human operators will most likely have important roles in delivery chains with ADV, such as loading and unloading of goods, as well as managing problems the ADV cannot solve. Consequently, how to design the human - ADV interactions will be critical from safety and efficiency points of view.

Publisher
p. 19
Series
Trafikverkets forskningsportföljer
Series
RISE Rapport ; 2022:131
Keywords
Automated delivery vehicles, Human – ADV interactions, Use cases, High-level functional requirements
National Category
Transport Systems and Logistics
Research subject
FOI-portföljer, Strategiska initiativ
Identifiers
urn:nbn:se:trafikverket:diva-16255 (URN)978-91-89757-20-2 (ISBN)
Projects
GLAD - Godsleverans under den sista milen med självkörande fordon
Funder
Swedish Transport Administration, TRV 2020/26017
Available from: 2024-08-19 Created: 2024-08-19 Last updated: 2025-09-04
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-2854-1477

Search in DiVA

Show all publications