Privacy Notice

Welcome to the ARRIVAL Platform!

We put great importance to data protection and therefore use the data you provide to us with upmost care. You can handle the data you provide to us in your personal dashboard. You will find our complete regulations on data protection and clarification of your rights in our legal notices. By using the website and its offers and navigating further, you accept the regulations of our privacy notice.


Challenge / Goal

The WINNER of the ENTRANCE Open Competition 2021 and of the special ENTRANCE prize!

The challenge of sustainable last mile logistics is similar to sea freight before sea container standards. The growing need and volume has produced a wide range of isolated approaches for the market. Unfortunately, they do not aim at unified resource sharing and access to diverse services. 

Vehicle-centric solutions: The current offering is mainly based on vehicles with a fixed cargo space (eBike, eScooter, eVan,  mobile robot, trailer). This practice prevents the cargo space from being automatically packed, consolidated and sorted into optimally sized batches even before the metropolitan area and city hubs with better facilities such as automatic sorters and packaging robots.

Lack of encapsulated cargo units prevents from using  commuter trains, metros and trams to bring goods and take massive amount of waste to/from the city. 

Narrow range of use: Current solutions mainly support only one type of freight transport, such as parcel delivery. Similarly, resources should be able to transport waste, recycled and building materials, machinery, furniture, sanitary facilities, refrigerated food, medicine, raw materials, groceries, breweries and food products to restaurants.                      

Fixed loading bays – resource bottleneck and misplacement: Unloading and loading of smart containers from/to vehicles require fixed docks. In a congested city and arbitrary places, it necessary to use plain street surface, such as curbsides for (un)loading.      

Lack of standards: Pooling and optimisation of resources fails due to missing common denominators. 
Deadlock in Autonomy: Existing solutions cannot later be developed into autonomous ones, making current solutions short-sighted and outdated very soon. Any new solution architecture must be future proof to enable current and future equipment to be utilized simultaneously. 



Our goal in general was to perform real-life technical and usability evaluation of the CONTAI system-level last mile solution based on the principle ’THE LEAST COMMON DENOMINATOR’, secondly to receive feedback and to make findings for further development.

We were willing to get feedback how the cross-use of different modules really work and could diverse use cases with appropriate applications be beneficial. Could combined door-to-door and locker pick-up work, could an AMR robot handle same applications as a pedal-assisted trike to contribute simultaneous use and smooth transition to autonomous last mile?

As preparing to autonomy is particularly important to secure a continuum for intelligent future, therefore special evaluation focus was in the deviation-tolerant and digitally assisted loading and unloading. Automatic operation is meant to be simultaneously used within manned and unmanned vehicles.

The development of last mile logistics is not only searching for suitable vehicles and routes but improvement of all process steps such as sorting, packaging and multimodal transfer of cargo from one vehicle to another, deliver cargo units to un|loading bays without any auxiliary. 

A sparsely facilitated ground-level city hub could scale, multiply and relocate quickly on demand basis. City events, city calendar, weather, citizen behavior, BigData, AI and the demand can steer the hub and mobile locker swarm placement and size. Ground-level hubs are possible if cargo units can be loaded and unloaded directly at street level, such as with the CONTAI technology.




CONTAI technical solution consists of modules and stable interfaces between them. 
Reading order: From bottom to top.

4 - Cargo Application, diverse, mounted on a Load Carrier.
3 - Load Carrier, optionally with IoT, variable lengths
2 - Loader mounted on the vehicle, electric, variable lengths
1 - Cargo Vehicle, diverse, manned and unmanned
0 - Ground / street



The implementation included several modules: 

Implementations of Load Carriers

The implementation included two Load Carriers:

1.  LoadCarrier 1:  L = 80 cm, base model 
2. LoadCarrier 2: L = 120 cm, with automatic brakes and a push handle with the dead man’s brake


Implementations of Loaders

The implementation included two Loaders:

1.  Loader 1:  L = 120 cm, 12V 150W,  200 kg, light model
2. Loader 2: L = 120 cm, 12V 625W, 800 kg, heavy model allowing multiple LoadCarrier lengths, e.g. 120...300 cm


Implementations of Cargo Applications

The implementation included seven different cargo applications mounted on top of the  Load Carrier:

1.  Hybrid locker
2. Recycle bin
3. Cold box
4. Irrigation water tank
5. Lattice gage 
6. Cassette for two roll cages
7. Container


Implementations of last mile vehicles

The implementation included four different light OEM vehicles with mounted loaders: 

1. Trike bicycle
2. Autonomous mobile robot - AMR
3. Bicycle trailer
4. Car trailer


1. The Load Carrier implementation (FIG 2)

Independent and at best unmanned loading and unloading of the load unit is strategically important in the effort to build a common last mile solution utilizing encapsulated cargo and equipment units (ref. sea container) that can be used anywhere. Also, e.g. when transferring the beverage deliveries of restaurants indoors the entire vehicle cannot get inside. Therefore heavy beverage containers must be unloaded on the street nearby. Carrying them inside is overwhelming and therefore a wheeled load carrier is needed.  Often there is no forklift or access to the loading dock is blocked - if there is one. A similar situation is not uncommon.

The common interface between the Load  and the Loader implements the CONTAI principle of 'The Least Common Denominator'. It allows the load unit to be separated from the vehicle in the same way as a sea container and a EUR pallet. It also gives a great deal of freedom to the vehicles used, both current and future. Likewise, it gives very high applicability to cargo units, the container is just one example and that too can come in different sizes. Movable toilet for disabled is not any standard container!

One of the implemented LoadCarriers is 120 cm long. Its guide rail width is a standard 55 cm and a total width of 80 cm. It can be fitted with wider applications (80 ... 260 cm). The LoadCarrier can also be longer, e.g. 300 cm, in which case min. 300 cm long Loader is required. The ground clearance of the Load Carrier is 10 cm, enabling forklift handling in accordance with the EUR pallet, and its total height is 13 cm (Figure 1). The Load Carrier has two fixed casters at the rear and two swivel wheels at the loading front end.

Qualified cargo unit parking requires the load unit to be left securely on the ground, i.e. it must not roll alone away on a sloping surface or in the wind. This is most important when the parking is performed by an unmanned robot. On the other hand, a wheeled load unit must be capable of being moved manually to perform last-foot service but loading and unloading require the wheels to rotate freely.

When loading the brakes are automatically off and when unloading takes end the brakes will be automatically locked. Parking brakes can manually be released with a dead man's handle like airport carts. Automatic parking brakes of LoadCarriers are necessary as unmanned commercial vehicles become more common, but they also speed up and simplify manual cargo unit handling.

The load capacity of the LoadCarrier was rated at 800 kg. The maximum load tested for loading and transport was 550 kg.


2. The Loader implementation (fig 3)

The CONTAI Loader and the LoadCarrier are designed so that they form an interface pair. The interface design allows universal Load Carriers to be loaded, lashed and transported completely autonomously. The Loader is attached to a carriage which moves on linear guides so that the Loader and the load can be electrically transferred longitudinally in the vehicle. This is advantageous in order to achieve an optimal weight distribution. The CONTAI Loader was attached to the carriage with a ball joint, allowing it to turn around all 3 axles during loading. The Loader is tapered at the rear end, which allows deviation-tolerant docking. The carriage is an electric motor driven and it has an automatically operating hook to grip the Load Carrier. The CONTAI Loader is 55 cm wide. The prototype was able to load 120 cm long Load Carriers. The proximity sensors at the rear part of the AMR robot triggers the docking sequence. 


3. The CONTAI HybridLocker application implementation (fig 4)

One experimental e-commerce locker mounted on the CONTAI LoadCarrier was built.

The CONTAI HybridLocker could combine doorstep deliveries with stand-alone locker pick-ups.
Once filled, the HybridLocker will be loaded on ground level onto a suitable vehicle. The HybridLocker makes first a door-to-door tour and it is then left close the distribution area at the curbside, garage, office lobby or even indoors for customer grocery and parcel pick-ups and recyclable returns without any hassle.
Conceptually the number of mobile lockers is big and their rotational frequency is high to minimize the pick-up distance and the use of private vehicles for that.

This new e-commerce hybrid process could allow customer to choose between synchronous and contactless, or asynchronous e-commerce services.
At the same time, the amount of unclaimed and missed deliveries, waste food and empty kms would decrease while recycling would increase.

3 The Autonomous cargo robot AMR implementation (fig 5)

Encapsulated cargo unit transport includes e.g. automatic docking, loading, lashing, transporting, unloading and parking and protection against vandalism and theft.

One functional AMR prototype was built during the implementation. The focus was in deviation tolerant LoadCarrier docking, loading, lashing, unloading and parking. Because the lack of the autonomous driving technology, autonomous driving for docking was simulated by a remote controlled radio sender - receiver. 

The AMR was able to autonomously load, lash and unload LoadCarriers with cargo applications, such as the HybridLocker.


How were end users involved in the development?

Several demonstrations with feedback gathering


Want to learn more about the lessons learned, financial details and results?

Log in

Time period

Planning time: 1 to 2 years

Implementation time: 6 months to 1 year

Something went wrong on our side. Please try reloading the page and if the problem still persists, contact us via
Action successfully completed!