December 01, 2015      

Starship Technologies, an Estonian startup led by Skype co-founders Ahti Heinla and Janus Friis, has created a six-wheeled, self-driving robot designed to solve the infamous “last-mile problem” — how best to deliver small packages to suburban homes.

Here’s the concept: Starship‘s robot can carry up to 40 lb. of goods, such as groceries, product samples, drugstore items, fast food, and small packages. It has a range of up to 2 miles on an electric motor that consumes 50 Watts.

The knee-high robot will be based in a modular warehouse (or retail outlet) for loading, unloading, and recharging before making its way out into the world.

Once out in the real world, the robot uses preprogrammed maps, GPS, and the public sidewalk to make its way to the delivery address. Obstacle- and collision-avoidance technologies protect it from “negative encounters” with pedestrians. Starship’s autonomous robots can move at 4 mph.

The robot is also fitted with an alarm, so any attempt to interfere with the device can be quickly brought to the attention of someone at the warehouse, supported by a high-definition video feed. Speakers and microphones enable the robot or the human worker to communicate with any pedestrians if necessary.

For “10 to 15 times less than the cost of current last-mile delivery alternatives,” the robot will arrive at its destination within five to 30 minutes of leaving the warehouse, according to Starship.

The customer can track the robot’s location in real time using a smartphone app and can use the same app to get access to the items he or she has been waiting for.

Once the delivery is complete, the robot makes its way back to the warehouse to prepare for its next journey.

Starship’s founders want to make local deliveries “almost free,” according to the company’s launch press release.

“Our vision revolves around ‘three zeroes’ — zero cost, zero waiting time, and zero environmental impact,” said Heinla, Starship Technologies CEO. “We want to do to local deliveries what Skype did to telecommunications.”

“For businesses, Starship’s technology eliminates the largest inefficiency in the delivery chain, the last mile,” the company said. “Instead of expensive and time-consuming door-to-door delivery, retailers can ship the goods in bulk to a local hub; then the robot fleet completes the delivery to the shopper’s door for a fraction of the cost.”

Starship Technologies is currently testing and demonstrating prototypes, and it plans to launch the first experimental delivery systems in England and the U.S. next spring. Commercial service is expected to launch in 2017.

Delivery limitations and challenges

“The system is not intended for crowded urban environments,” Heinla acknowledged to The New York Times. “Rather, it is targeted for relatively affluent and uncrowded suburban neighborhoods, gated communities, assisted living facilities, and campuses, where it will travel on sidewalks, programmed to mingle freely with pedestrians, bicyclists, and cars.

So don’t expect to see delivery robots trundling all over busy city-center sidewalks anytime soon. These are designed to be suburban robots, not city slickers.

There are legal and regulatory challenges, too.

The obvious legal issue deals with safety and personal injury, said Garry Mathieson, an expert in robot law and co-chair of the Robotics, AI, and Automation Industry Group at Littler Mendelson PC.

“Many local communities have sidewalk laws primarily focusing on bicycles but also often covering four-wheel vehicles like golf carts,” he said. “I suspect that as Starship becomes more common, local communities will amend their sidewalk laws to include ‘ground drones.'”

“If, for example, a delivery robot collides with a small child, personal injury claims will arise,” explained Mathiason. “The question would be negligence and foreseeability.”

Another area involving legal controls is using delivery robots to transport hazardous materials, something Mathiason does not recommend until the system has been proven safe.

Jurisdiction over public sidewalks is predominantly given to local governments, said Mathiason.

“If the robots become popular and significantly slow pedestrian traffic or bicycles — where they are permitted — local ordinances are likely limiting the use of delivery robots and/or restricting the hours of operation,” he said.

The legal issues are challenging, but far from insurmountable.

“There are several other legal issues which will be created by delivery robots, but they can be answered,” said Mathiason. “If the system is effective and substantially reduces delivery time, the legal challenges will find solutions.”

He noted that last-mile delivery robots are likely to require licensing, which will almost certainly be accompanied by a series of safety and operational requirements.

Similarly, Allan Martinson, Starship’s chief operating officer, observed that that new technologies often require regulatory changes. He pointed out that the Segway two-wheeled motorized transportation vehicle has received such exemptions in many locations.

Solving the last-mile problem

Before we take we look at Starship’s competition, let’s delve a bit deeper into last-mile logistics.

“Modern transportation networks generally consist of a ‘hub-and-spoke‘ model: Between the hubs, transport is very efficient due to the clustering of products which creates high-volume flows (line hauls),” according to ORTEC BV, a multinational optimization software and analytics provider.

“However, the ‘feeding networks’ (or first mile) and ‘local distribution’ (or last mile) to/from the spokes are typically more costly,” said ORTEC. “The main reason is that it’s not possible to have the same level of clustering due to the need to do door-to-door deliveries. This results in low volumes to do deliveries and pickups at individual customer locations.”

That last mile is particularly expensive and problematic for two main reasons. First, traditional methods involve a high degree of failed deliveries, returns, and empty vehicles.

Second, delivery vehicles typically tend to be small vans, which are relatively expensive to operate and pose environmental issues.

Are last-mile delivery robots the solution to a problem we didn’t know we had?


Two key trends are driving interest in last-mile logistics: the rapid growth of online sales and the growing number of working hours being put in by the average worker, which makes it increasingly difficult to complete a home delivery at the first attempt.

So, last-mile delivery robots such as Starship’s address a real and growing need.

Furthermore, they promise to do so at a dramatically reduced cost compared with traditional delivery methods. Starship’s target is £1 ($1.51) per delivery.

Sidewalk competition

Direct competition for Starship Technology’s robot includes the Sidewalk delivery robot. It promises many of the same benefits as Starship in terms of efficiency and speed.

Sidewalk would be remotely operated and can travel up to 7 hours on a single charge.

Not only does Lithuania-based Sidewalk have a prototype delivery robot, but it has also partnered with DHL Express to collaborate on future research.

The company emerged from Startup Sauna, a startup accelerator program held two months ago in Finland. Sidewalk has already tested deliveries for cafes, flower sellers, and e-food delivery platforms.

Meanwhile, U.S.-based startup Dispatch Network was founded by two American participants in a Chinese hardware accelerator. According to reports, the company already has a basic prototype that’s larger than Starship.

Drone rivals

All of these startups will face stiff competition from another robotics technology that is seen as a solution to the last-mile problem: unmanned aerial vehicles (UAVs) or drones.

Major technology players and retailers are involved in research and development of airborne delivery. Inc. has been developing Amazon Prime Air since at least 2013. The online retailer claims that its drones can fly 2.3 kg (about 5 lb.) of goods to customers within 30 minutes of an order being placed. Amazon has delivery drone development centers in the U.S., the U.K., and Israel.

In early November, Google Inc. announced that it intends to launch its Project Wing drone delivery business by 2017. The company has completed preliminary tests of its drone delivery system in Australia.

And retail giant Walmart plans to use DJI-manufactured drones for package pickup and delivery and for checking warehouse inventories.

Drone deliveries are being tested in Australia, Finland, Singapore, and the U.S.

Terrestrial delivery robots offer some advantages over drones, however. Accidents are less likely to have terrible consequences when one compares a slow-moving sidewalk-based robot to the prospect of a UAV falling out of the sky.

Terrestrial robots are able to carry larger payloads, and operate for a longer time. The mobile technology is similar to that already used in warehouses. Regulators will likely react soon to both aerial and ground-based drone deliveries.

Further robotic competition is likely to come from self-driving cars, with many seeing these vehicles as the long-term solution to last-mile challenges.

Non-robotic competition

In addition, non-robotic solutions have been proposed for deliveries, but they have met with mixed success.

For example, Canada Post, announced that it would no longer provide door-to-door delivery of letter mail in some regions a few years ago. People had to make their way community mailboxes instead.

It’s a strategy retailers have employed too, with both Walmart and Best Buy encouraging customers to pick up their packages at nearby stores.

Amazon has installed metal storage lockers at convenience stores and drugstores in various U.S. cities where customers can pick up their purchases from storage lockers.

The notion of “consolidated deliveries” has also emerged in recent years. Instead of having empty vehicles making deliveries in dense urban areas, consolidation centers can ensure that only fully-loaded vehicles take on the last mile of the journey.

For now at least, the non-robotic competition appears light and more like a temporary stop-gap on the way to the fully automated supply chain that customers and suppliers are crying out for.

Starship Technologies preps for launch

The prospects for companies like Starship look good. The company is looking for an initial round of venture capital funding.

Consumer and supplier demand for a solution to the last-mile problem is high. The legal issues associated with delivery robots on our sidewalks can be resolved in time.

And the technology is capable: Starship’s robot is made from off-the-shelf components, so the cost of entry is low for developers.

The biggest challenges are likely to come from driverless cars and, to a lesser extent, drones.

Starship Technologies’s co-founder Heinla hit the nail on the head when he observed that the robot is best suited to quiet, suburban streets, not bustling cities. This is a limitation, not just in terms of capabilities, but in terms of the potential market for these devices.

Nevertheless, it seems certain that last-mile delivery robots will find niche applications in gated communities, sports clubs, universities, and other low-traffic environments.

More on Drone Applications: