Automating the Final Frontier: Veho and Rivr Tackle the 'Last 100 Yards' of Delivery

In the complex world of logistics, automation has revolutionized many stages of the supply chain, from warehouse management to long-haul transportation. Yet, a persistent challenge remains: the final leg of the journey, often referred to as the 'last mile.' This segment, encompassing the delivery from a local hub or vehicle to the customer's doorstep, is notoriously inefficient and costly. While many companies have explored automating aspects of the last mile, most solutions stop short of the final, most intricate steps – navigating sidewalks, driveways, and crucially, stairs.

This is precisely the challenge that logistics company Veho and Zurich-based robotics startup Rivr are setting out to solve. Recognizing that the real bottleneck and opportunity lie in those critical final 100 yards from the delivery van to the customer's front door, the two companies have launched a pilot program in Austin, Texas. This collaboration introduces Rivr's unique four-wheeled, stair-climbing delivery robot into Veho's delivery operations, aiming to bridge the gap between curb and doorstep autonomously.

The 'Last 100 Yards': A Unique Challenge in Logistics

The concept of the 'last mile' is well-established in logistics, representing the final leg of a package's journey. It's often the most expensive and time-consuming part of the delivery process due to factors like traffic congestion, parking difficulties, navigating varied urban and suburban environments, and the sheer number of individual stops required. Automating the last mile has been a significant focus for innovation, leading to the development of sidewalk delivery robots, drones, and autonomous vehicles.

However, even these innovations often address only a portion of the problem. Sidewalk robots, for instance, can handle flat, predictable paths but struggle with obstacles like curbs, uneven terrain, and stairs. Drones face regulatory hurdles and limitations in payload and range. Autonomous vehicles are still primarily focused on the road segment. The 'last 100 yards' – the distance from where a vehicle can safely stop to the actual delivery point – presents a distinct set of challenges that require a different approach.

This final segment involves navigating potentially complex and unpredictable environments: walking up driveways, crossing lawns, opening gates, and, critically, ascending and descending stairs. These tasks are trivial for a human delivery driver but pose significant technical hurdles for conventional robots. Marko Bjelonic, CEO and founder of Rivr, highlights this point, stating that in last-mile delivery, "robotics makes an impact by actually solving these very challenging problems that are actually quite easy for humans but hard for robots."

The ability to reliably and safely traverse stairs is a key differentiator for Rivr's robot. Many residential and even some commercial delivery locations involve steps, making a robot that can handle them essential for true doorstep delivery automation. By focusing on this specific, difficult problem, Rivr aims to provide a solution that complements existing last-mile automation efforts, effectively acting as the "next evolutionary step from the sidewalk robots."

Introducing the Rivr Robot: A 'Dog on Roller Skates'

Rivr's robot, described by Bjelonic as looking like "a dog on roller skates," is designed specifically to tackle the complexities of the final few feet of delivery. While the exact technical specifications are not fully detailed, its four-wheeled design and reported stair-climbing capability suggest a sophisticated locomotion system capable of adapting to varied surfaces and inclines. Unlike robots that rely solely on tracks or multiple legs, a wheeled system with advanced suspension and articulation can offer a balance of speed on flat ground and capability on uneven terrain and stairs.

The robot's design likely incorporates advanced sensors – cameras, lidar, ultrasonic sensors – to perceive its environment, detect obstacles, and map its path. Navigation in the 'last 100 yards' requires precise localization and mapping, often in areas without detailed pre-existing maps. The robot must be able to identify the correct delivery location, find the optimal path to the door, and safely navigate any intervening obstacles, including steps.

Safety is paramount for robots operating in public or semi-public spaces. The Rivr robot must be able to detect and avoid people, pets, and unexpected objects. Its relatively small size, perhaps suggested by the 'dog' analogy, could contribute to its maneuverability and reduce potential hazards compared to larger autonomous vehicles.

The Austin Pilot Program: Testing in the Real World

The partnership between Veho and Rivr is kicking off with a focused pilot program in Austin, Texas. Austin, known for its mix of urban density and suburban residential areas, provides a suitable testing ground for a robot designed to handle varied environments, including those with stairs.

The initial phase of the pilot is intentionally small-scale. According to the companies, just one highly supervised robot will be deployed daily. This cautious approach allows Veho and Rivr to closely monitor the robot's performance, identify potential issues, and gather data in a controlled environment. The robot will operate for five- to six-hour runs over a couple of weeks, focusing initially on the more residential areas of Northwest Austin before potentially expanding to denser parts of the city.

Human supervision is a critical component of this early pilot. A Rivr employee will accompany the robot during its runs to ensure safety and delivery quality. While the bots are designed to operate autonomously, remote operators will also be available to take control if the robot encounters a situation it cannot resolve on its own, such as getting stuck or facing an unexpected obstacle. This human-in-the-loop approach is common in early autonomous system deployments, providing a safety net and valuable learning opportunities.

For Veho, a company that provides delivery services across 50 U.S. markets for major brands like Sephora, Saks, and HelloFresh, this pilot is a crucial test of automating the final steps of delivery. Veho already focuses on a customer-centric delivery experience, and integrating robotics could potentially enhance this by increasing efficiency and freeing up drivers for other tasks.

Motivation and Mutual Benefits: Data, Efficiency, and Reduced Workload

The partnership offers significant benefits for both Veho and Rivr, addressing key challenges and opportunities in the autonomous delivery space.

For Rivr: Data Collection and AI Development

For Rivr, the pilot program with Veho is a vital opportunity to test its technology in real-world delivery scenarios and, perhaps more importantly, to accumulate the vast amounts of data necessary to build a robust and general physical AI framework. Developing AI for physical robots operating in unstructured environments is fundamentally different from training language models or autonomous vehicles.

Marko Bjelonic points out the disparity in available training data: "ChatGPT and other chatbots have the internet as training data, and autonomous cars have thousands of cars on the street that they can attach sensors to and start collecting data." In contrast, he notes, the robotics world lacks such massive, readily available datasets for complex physical tasks like navigating diverse residential properties.

Therefore, finding meaningful use cases where robots can solve real problems is essential for collecting the necessary data to make them more intelligent and capable. The Veho partnership provides Rivr with access to a controlled yet realistic environment to gather data on navigation, obstacle avoidance, interaction with the environment (like stairs and doorsteps), and handling packages. This data is invaluable for refining Rivr's algorithms, improving the robot's autonomy, and potentially developing more generalized robotic capabilities.

For Veho: Enhancing Efficiency and Supporting Drivers

For Veho, the pilot represents a chance to explore how automation can extend beyond the van. Automating the 'last 100 yards' could unlock new levels of efficiency, particularly in dense urban or suburban areas. Fred Cook, Veho's co-founder and CTO, envisions scenarios where a driver and a robot could tag-team a street simultaneously, with the robot handling deliveries to doors while the driver focuses on other tasks or deliveries requiring human interaction.

This division of labor could potentially allow drivers to make more deliveries in a given timeframe, increasing overall throughput. Furthermore, automating the physically demanding task of walking door-to-door, especially when navigating stairs or carrying heavy packages, could significantly reduce the workload on human drivers. Bjelonic suggests that Rivr's robots can act as "robot helpers" in this regard, improving driver well-being and potentially reducing delivery times.

The pilot also allows Veho to understand the operational requirements of integrating robots into their existing workflow. Cook mentions the possibility of pairing robots with certain types of vehicles equipped with charging stations, allowing the bots to operate for a full day without needing to return to a central depot for power. This kind of operational insight is crucial for scaling the technology in the future.

The Broader Context: The Future of Autonomous Delivery

The Veho-Rivr partnership is part of a larger trend towards increasing automation in the logistics industry. As e-commerce continues to grow, the demand for faster, cheaper, and more efficient delivery solutions is escalating. The last mile remains the most significant cost center in delivery, making it a prime target for technological disruption.

While fully autonomous, end-to-end delivery is still some way off, initiatives like this pilot program demonstrate a focus on automating specific, challenging segments of the process. By tackling the 'last 100 yards,' Veho and Rivr are addressing a gap that many other autonomous delivery solutions have yet to fully bridge.

The development of robots capable of navigating complex, unstructured environments like residential properties is crucial for the widespread adoption of autonomous delivery. Sidewalk robots have faced challenges with public acceptance, regulations, and physical obstacles. Robots that can operate closer to the delivery vehicle and handle varied terrain might offer a more practical and scalable solution for certain types of deliveries.

Furthermore, the emphasis on data collection in the Rivr-Veho partnership highlights the iterative nature of AI and robotics development. Real-world deployment and the data it generates are essential for training and improving autonomous systems. Each delivery attempt, each stair climbed, and each obstacle encountered provides valuable information that can be used to refine the robot's perception, navigation, and decision-making capabilities.

Rivr's Growth and Future Ambitions

Rivr, also known as Swiss-Mile, is not a newcomer to the robotics scene. The startup has been developing its unique mobile robot technology and has already secured significant funding, including participation in a round led by Jeff Bezos that valued the company at $100 million. This level of investment underscores the perceived potential of their technology to solve difficult problems in logistics and other industries requiring advanced mobile manipulation.

Beyond the Austin pilot, Rivr is already operating in the U.K. through a partnership with delivery platform Evri, indicating their technology is being tested and deployed in multiple markets. The company has ambitious scaling plans, hoping to increase its fleet to 100 bots by next year and thousands in 2027. Achieving these goals will depend heavily on the success of pilot programs like the one with Veho, which provide the necessary data and operational experience to refine the technology and demonstrate its value proposition.

Veho's Strategic Vision for Automation

Veho's participation in this pilot aligns with its broader strategy of leveraging technology to optimize its delivery network. Veho differentiates itself through its focus on a high-quality, customer-friendly delivery experience, often providing scheduled and tracked deliveries directly to consumers. Integrating robotics for the final steps could further enhance this service by increasing reliability and potentially offering more precise delivery timing.

Veho previously raised a significant funding round, securing $170 million in Series B funding in early 2022. This investment has fueled its expansion across the U.S. and its ability to invest in technological advancements like the Rivr partnership. By exploring automation solutions for the most challenging parts of the delivery process, Veho aims to stay ahead in the competitive last-mile delivery market.

Challenges and the Road Ahead

While the Austin pilot is a promising step, significant challenges remain for widespread adoption of robots like Rivr's. Regulatory frameworks for autonomous robots operating in public spaces are still evolving. Public perception and acceptance of robots on sidewalks and properties need to be considered. The cost of manufacturing and maintaining these sophisticated robots must be balanced against the potential efficiency gains.

Furthermore, the technical challenges of navigating highly variable and unpredictable environments cannot be overstated. While the robot can handle stairs, it must also contend with weather conditions, unexpected obstacles, varying property layouts, and potential interactions with humans and animals. The human supervision in the initial pilot is a recognition of these complexities.

However, the structured nature of the pilot, starting in residential areas and gradually expanding, allows Veho and Rivr to address these challenges incrementally. The data collected will be crucial for improving the robot's autonomy and reliability, eventually reducing the need for direct human supervision.

Conclusion

The partnership between Veho and Rivr represents a focused effort to automate one of the most persistent and challenging segments of the delivery process: the 'last 100 yards' from the van to the doorstep. By deploying Rivr's unique stair-climbing robots in a real-world pilot in Austin, the companies are not only testing the feasibility of this technology but also gathering essential data to advance the field of physical AI in robotics.

For Veho, this initiative offers the potential to enhance efficiency, increase delivery capacity, and improve the working conditions for its drivers. For Rivr, it provides a critical use case for its technology and a pathway to collect the data needed for scaling its operations and improving its robots' intelligence. While the pilot is starting small and under close supervision, it marks a significant step towards a future where autonomous robots play a more integral role in completing the final, complex steps of parcel delivery, potentially transforming the economics and logistics of the last mile.

The success of this pilot could pave the way for wider deployment of such robots, bringing us closer to a future where packages are reliably and efficiently delivered directly to our doors, regardless of the steps or obstacles in the way. It underscores the ongoing innovation required to solve the intricate challenges of modern logistics and the increasing role of specialized robotics in achieving true end-to-end automation.