Delivery Robots Future: Advantages, Challenges, and Implications

20 Min Read

The rise of delivery robots is a hot topic in the logistics industry. These robots have the potential to revolutionize the way we get our goods, but there are also concerns about their impact on human jobs.

One of the biggest advantages of delivery robots is that they can operate 24/7, which means that deliveries can be made more quickly and efficiently. They are also less likely to make mistakes than human drivers, and they can navigate difficult terrain that would be unsafe for humans.

However, there are also some potential drawbacks to delivery robots. One concern is that they could lead to job losses in the logistics industry. Another concern is that they could be hacked or vandalized. Overall, the future of delivery robots is uncertain. There are both potential benefits and risks associated with these robots. However, it is clear that they have the potential to change the way we get our goods.

Autonomous last-mile delivery has gained momentum, particularly during the pandemic, due to the need for safe and contactless deliveries. The technology offers several advantages, including safe delivery of goods, sustainability, overcoming inaccessibility, predictable delivery times, regulatory acceptance, reduced uncertainties related to manual labor, and technological advancements. The global autonomous last-mile delivery market is projected to grow significantly in the coming years.

This trend is expected to continue growing for several reasons:

  1. Safe delivery of goods: Customers preferred autonomous delivery vehicles and bots to minimize the risk of disease transmission from human delivery agents. Contactless authentication methods, such as scanning QR codes, further enhance safety.
  2. Sustainability: Autonomous vehicles (AVs) used for delivery are often electric-powered and can utilize renewable energy sources for charging. Smart software systems optimize routes, reducing energy consumption and emissions, aligning with sustainability goals.
  3. Overcoming inaccessibility: Autonomous delivery systems provide solutions to deliver goods to remote regions using technologies like drones and ziplines. AVs and robots, with their smaller size and agile navigation capabilities, can navigate urban traffic more efficiently, reaching destinations faster.
  4. Predictable and accurate delivery time: Software-suggested routes and continuous tracking of autonomous deliveries enable more precise estimates for delivery times, improving customer satisfaction.
  5. Regulatory acceptance: While regulations for fast self-driving cars on public roads are restrictive, many regulatory bodies are more open to considering autonomous last-mile delivery systems. These systems are generally small, carry smaller payloads, and have lower potential for harm in case of accidental contact with humans.
  6. Reduced uncertainties related to manual labor: Autonomous delivery systems eliminate challenges associated with manual labor, such as labor availability and pricing fluctuations, providing more reliable service.
  7. Technological advancements: Advances in sensing technologies, mapping, artificial intelligence (AI), and cloud-based deployments have contributed to the growth of autonomous delivery systems. There is a shift towards utilizing self-driving technologies in smaller, immediate use-cases.

Market projections indicate significant growth in the autonomous last-mile delivery sector. According to Fortune Business Insights, the market is expected to reach $51.38 billion by 2028 with a compound annual growth rate (CAGR) of 24.4%. Allied Market Research estimates that the market could reach approximately $90 billion by 2030.

McKinsey’s report highlights the potential of drones in the delivery supply chain, noting that over 2,000 commercial drone deliveries were taking place daily worldwide in early 2022. Drones offer advantages for last-mile deliveries, particularly in regions with poor road infrastructure. They are small, fast, and environmentally friendly, using less energy and producing fewer emissions than conventional means.

Drone delivery is subject to strict regulations due to security and privacy concerns, limiting their access to certain areas. Despite these challenges, major industry players like Skyports, Airbus, Matternet, Walmart, and Amazon have already ventured into drone delivery. Walmart, for example, has established 37 stores for drone delivery, partnering with DroneUp Delivery to transport packages weighing up to 4.5kg.

Self-driving trucks and vans have been utilized in off-road applications for industries like mining, agriculture, and construction, as well as for internal transportation within private campuses. Major truck manufacturers such as Volvo, Caterpillar, and Komatsu, along with startups like Aurora, Embark, Waabi, Waymo, etc. are now focused on achieving Level-4 autonomy for highway driving. Many of these companies claim that their technologies will be ready for deployment this year. Automating long-haul deliveries is a significant goal for truck manufacturers, as highway driving can be monotonous for human drivers compared to shorter local trips.

In a demonstration conducted by Kodiak Robotics and U.S. Express, a self-driving eighteen-wheeler successfully completed non-stop trips between Dallas and Atlanta, covering over 10,000 kilometers in just five days. This would have taken human drivers ten days with breaks. The trips were supervised by drivers who took turns behind the wheel, but interventions were necessary during the test.

Another category of autonomous trucks focuses on middle- and last-mile delivery, transporting goods between warehouses or local stores, which requires navigating public roads. Microsoft is planning to invest over $10 million in Gatik, a California-based company that advocates for autonomous truck usage on public roads. Gatik’s customers include Walmart, Georgia-Pacific, KBX, and Pitney-Bowes, and they employ medium-duty Level-4 ready trucks from Isuzu.

Udelv, another California-based company, specializes in middle- and last-mile delivery with its cabless autonomous vehicle called the Transporter. The Transporter features a modular cargo pod with a capacity of 900kg. Traveling at speeds of up to 112kmph, Udelv’s vehicle can cover distances ranging from 250 to 500 kilometers, depending on the battery pack, and make up to 80 stops per cycle. Udelv aims to have 50,000 vehicles operating on public roads by 2028.

Despite the potential of self-driving trucks, they are heavily regulated due to their weight and potential hazards in case of malfunctions. They are currently permitted only in select regions worldwide. According to Richard Bishop, a Forbes contributor and trend watcher, 2024 seems to be the target year for commercial deployment of autonomous trucks on highways, with 2023 focused on conducting final on-road validation testing. The launch dates from original equipment manufacturers (OEMs) are closely guarded, but it is unlikely that they will make a move before then.

Ground delivery bots and small autonomous vehicles have gained an advantage in the autonomous delivery industry due to their smaller size and the greater acceptance by regulatory bodies compared to large trucks. These bots and vehicles, typically equipped with four or six wheels, operate on sidewalks or streets to deliver food and parcels to end consumers. With the help of sensors, they can assess obstacles and navigate safely around people, animals, and objects. These vehicles are compact, move at slower speeds, and have limited payload capacities.

Restaurants, grocery stores, and other businesses partner with service providers that offer app-based robots for deliveries within a localized radius. When a customer places an order, a robot from the network is assigned to pick up the items from the vendor and deliver them to the designated location. The robot often has a locker box to securely hold the items for delivery, and the customer receives a password or uses contactless authentication to access the locker box and retrieve their order.

Autonomous last-mile delivery robots come in various sizes and functionalities. Some, like Nuro, resemble small cars, while others are compact and visually appealing, resembling locker boxes on wheels. They may operate on pavements or roads, and their capabilities range from single-delivery bots to those with modular containers capable of managing multiple stops.

Amazon was one of the pioneers in this field, conducting a nearly three-year trial of their six-wheeled delivery device called Scout in California. However, they scaled back the project last year. FedEx also developed an intriguing same-day delivery robot called Roxo in collaboration with renowned inventor Dean Kamen. Roxo, an eye-catching robot with flashy lights, was even capable of climbing curbs and steps to deliver packages directly to customers’ doors, making it more accessible for individuals with limited mobility. However, FedEx discontinued the project last year to focus on other immediate opportunities.

Nonetheless, these early adopters served as inspiration for numerous startups, leading to a highly competitive landscape in the autonomous last-mile delivery sector today.

Nuro was granted the first autonomous exemption from the National Highway Traffic Safety Administration in the USA. Their four-wheeled machine weighs approximately 700kg and can reach a top speed of around 70kmph. It has a payload capacity of 225kg and features customizable compartments with heating and cooling capabilities. Nuro has conducted pilot programs in Texas, Arizona, and California, collaborating with partners such as Dominos, Kroger, Walmart, FedEx, and 7-Eleven. They recently signed a 10-year agreement with Uber Eats for robot food delivery.

Not all delivery robots are as large as Nuro. Companies like Starship, Kiwibot, Cyan Robotics, and Ottonomy.IO offer small and adorable robots for last-mile deliveries. Kiwibot has deployed over 50 robots across more than 25 universities and has completed over 200,000 deliveries in cities like San Jose, Santa Monica, Denver, Dallas, as well as parts of Taiwan and Colombia.

Coco, developed by Cyan Robotics, specializes in 15-minute food deliveries from local restaurants. On the other hand, Tortoise introduced mobile vending robots that vendors can load with goods, allowing the robots to sell items as they navigate campuses or street sidewalks. Customers can select their desired products and self-checkout, with the robots being remotely supervised by human operators.

Ottonomy.IO’s Ottobot focuses on delivering food near passengers’ gates at airports, operating in several airports including Cincinnati, Pittsburgh, and Rome. At CES, Ottonomy.IO unveiled Yeti, a self-dispensing robot capable of delivering goods even when the customer is not available at home to receive them.

Goggo Network, an autonomous mobility solutions provider in Europe, offers their services to multiple partners. They provide both small and large robots for single or multiple-stop deliveries. They participated in the French government-funded 5G Open Road project and currently operate in France and Spain, working with clients such as Carrefour and DIA.

Swiss startup LOXO plans to launch its mobile delivery robots on public roads this spring, starting in Switzerland and expanding to the UK and Germany. Their autonomous vehicle is a compact, electrically-powered box on wheels with compartments for packages. Equipped with various sensors, LOXO’s self-driving robot travels at typical urban traffic speeds and halts immediately upon detecting any potential danger. Trained personnel remotely monitor its operation and can assume control if necessary, while also being able to communicate with users.

Autonomous last-mile delivery bots share many similarities with self-driving cars, but they also require additional technology. These bots are equipped with various sensing technologies, including thermal cameras, point cloud sensors, lidar, radar, ultrasonic sensors, and more. They also feature processing units such as central processing units, graphics processing units, neural network accelerators, field-programmable gate arrays, and embedded integrated circuits.

The backbone of these autonomous systems is the software. Each bot has a complex software system with millions of lines of code, both onboard the robot itself and in the cloud. The robot’s software enables it to perceive its surroundings, avoid obstacles, detect objects, brake, accelerate, and navigate corners. It also communicates its location and reports any issues to central locations. The cloud software handles tasks like payments, route planning, fleet orchestration, robot health monitoring, energy management, and ensuring the right number of robots are in the right place at the right time.

The robots maintain a data connection, either cellular or Wi-Fi, to communicate with the cloud. In the event of a connection failure, the robots continue to operate safely within permitted routes until they regain connectivity.

During their “rest time,” the robots typically move to a hub for charging and software upgrades. Diagnostics data is uploaded to the cloud, allowing companies to analyze the robot’s performance and make improvements.

Managing a fleet of autonomous delivery robots is a significant shift compared to traditional autonomous vehicle operations. Operating a fleet of 10 vehicles versus 200 to 2,000 delivery robots brings educational benefits in areas like autonomous fleet management, distributed AV information and processing, fallback safety, and fleet orchestration. The increased deployment of delivery robots exposes them to various edge cases, such as challenging sidewalks, airports, and curbsides. These scenarios drive advancements in active perception, situational awareness, autonomous behaviors, and safety.

Design plays a crucial role in the development of delivery robots. The unique requirements of the delivery function shape the design choices. Storage is a key consideration, with options ranging from modular cargo boxes to simple locker boxes, depending on the robot’s size and purpose. Some robots feature compartments with heating and cooling capabilities. Additionally, delivery robots need to have a friendly and non-threatening design, characterized by soft, rounded corners. For safety, certain robots, like Nuro, even include airbags in the front to protect pedestrians in case of accidental contact.

Delivery robots must adhere to local rules and regulations, operating only in permitted areas. They are equipped with features like lights and alarms to signal their movements to pedestrians. Moreover, these robots are designed to be courteous and respectful. They are programmed to say “excuse me” when someone blocks their path and express gratitude with a “thank you” once they are allowed to proceed.

While many perceive autonomous delivery as a threat to human employment, it may not necessarily lead to job losses. Instead, it can relieve humans from monotonous tasks, challenging shifts, long work hours, and potentially hazardous conditions. This allows human workers to focus on more subjective and valuable tasks. An example of this can be seen in the automotive industry, where increased automation has not resulted in a significant reduction in human employment. In fact, the industry has seen growth in the number of workers over time.

At this stage of technological growth, it is challenging to halt progress. Therefore, it is more productive to seek a balance between automation and human labor. Unlike self-driving cars and heavy vehicles, autonomous last-mile delivery systems, especially mobile robots, are poised for significant growth in the near term. Observers predict that more countries will grant permissions for small and slow autonomous vehicles to operate on sidewalks, roads, and public spaces, which may be more accepting compared to faster vehicles. Over time, economic forces will help find a harmonious equilibrium between human-powered and machine-powered tasks.

Automated and Connected Vehicles

Automated vehicles use digital technologies to assist drivers with some or all driving functions. Self-driving or driverless vehicles are automated vehicles that can operate without human input. Connected vehicles are equipped with devices that allow them to communicate with other vehicles or infrastructure via the internet.

Levels of Automation and Timeline

Automated vehicles are classified into six levels of automation, from Level 0 (no automation) to Level 5 (full automation).

  • Level 0: No automation. The driver is fully responsible for all driving tasks.
  • Level 1: Driver assistance. The vehicle can provide limited assistance with driving tasks, such as lane keeping assist or adaptive cruise control.
  • Level 2: Partial automation. The vehicle can take control of some driving tasks, but the driver must still be prepared to take over at all times.
  • Level 3: Conditional automation. The vehicle can take control of all driving tasks under certain conditions, such as on highways or in traffic jams. The driver must still be prepared to take over if necessary.
  • Level 4: High automation. The vehicle can take control of all driving tasks, including parking and handling unexpected events. The driver is not required to be present in the vehicle.
  • Level 5: Full automation. The vehicle can operate without any human input.

Vehicles that assist drivers (Levels 1 and 2) are already on the market in Europe. Self-driving vehicles (Levels 3 and 4) are currently being tested and are expected to be available on the market between 2020 and 2030. Fully automated vehicles (Level 5) are expected to be available as of 2030.

Share This Article