2026 Agenda
-
October 18, 2026
-
October 19, 2026
-
October 20, 2026
-
October 21, 2026
-
3:00 pm-8:00 pm
Exhibitor Move - In
Event Set-UpWhereHall A & B
-
8:00 am-4:00 pm
Exhibitor Registration & Move-In
Exhibitor HallWhereHall A & B
-
8:00 am-4:00 pm
Registration Kick Off
RegistrationWhereMain Lobby Foyer
-
8:00 am-6:00 pm
Registration Hours
RegistrationWhereMain Lobby Foyer
-
8:30 am-9:30 am
Exhibitor Set Up & Move-In
Exhibitor HallWhereHall A & B
-
9:15 am-9:20 am
Opening Remarks & Special RoboBusiness Announcement
GeneralWhereTheatre - Floor 2
-
9:20 am-10:05 amOpening Keynote: The Next 20 Years of Robotics
Robotics has evolved from a promising technology into a critical business tool across manufacturing, logistics, healthcare, agriculture, defense, and beyond. As RoboBusiness marks its 20th anniversary, industry leaders reflect on the breakthroughs that defined the past two decades and explore the technologies, markets, opportunities, and challenges that will shape the next twenty years.
GeneralWhereTheatre - Floor 2
-
10:00 am-6:30 pm
Exhibit Hall Hours
Exhibitor HallWhereHall A & B
-
10:05 am-10:30 am
AM Networking Break
Breaks/MealsWhereHall A & B
-
10:30 am-11:15 amDesigning Data Programs for Humans, Not Robots
Skilled, high-variability work, such as a line cook cutting onions, a housekeeper turning a room in under fifteen minutes, or a warehouse worker depalletizing an inbound shipment, presents some of the hardest environments to capture good training data from. Aaron will talk through what that gap looks like in practice, what it takes to capture usable data from real workers on real shifts, and what he's learned putting camera systems into commercial environments at scale.
AI TrackWhereRoom 209/210
-
10:30 am-11:15 amStartup Bootcamp Part 1: The Basics to Best Practices
Starting a robotics company, as well as growing and scaling, has many challenges. Part 1 of the Robotics Startup Boot Camp will provide best practices that includes discussions around the many common challenges that can make or break a company. Experts will speak about topics that include:
-
Intellectual Property & Trademark Considerations
-
How to Fund Your Startup
-
RaaS Agreements and Structure of Contracts
Business & Deployment TrackWhereRoom 207
-
-
10:30 am-11:15 amDesigning the Robotics Ecosystem for Outdoor Autonomy
Mel Torrie, CEO and Founder of Autonomous Solutions Inc. (ASI), is a global leader in vendor-agnostic vehicle automation. With over two decades of robotics leadership, Mel has pioneered autonomous technologies across the mining, agriculture, construction, and logistics industries.
In this forward-looking session, Mel will explore why the intersection of autonomous vehicles and advanced robotics will further revolutionize field deployments across. While today’s autonomous vehicle fleets successfully address pain points in geographically remote or rugged locations, improving worker safety and operational efficiency, the next ten years will see a massive leap forward. The use of mobile robotics in outdoor applications is poised to advance significantly over the decade. To unlock the next level of productivity, autonomous vehicle solutions must be intentionally designed to support that integration and foster a collaborative relationship between platforms.
Mel will share insights on how building a unified, cooperative ecosystem where physical robotics seamlessly accompany and work alongside autonomous vehicles will compound operational benefits. Attendees will gain a strategic roadmap for navigating this shifting landscape, learning how the symbiosis of rugged vehicle platforms and intelligent robotics will redefine predictable productivity and safety in the world's most challenging work environments.Field Robotics TrackWhereRoom 206
-
10:30 am-11:15 amThe Humanoid Bill of Materials: De-Risking the Multi-Billion Dollar Bottleneck
Featuring Michelle Sun (Founder, Core Matter Labs)
The humanoid race will not be won purely by novel foundation models or elegant software demos. It will be decided by who controls the physical infrastructure down the stack. While the Western technology ecosystem remains hyper-focused on the intelligence layer, the actual scaling and deployment bottleneck sits squarely within highly concentrated component tiers. Actuators and precision mechanical reducers alone command 50% to 65% of a humanoid robot's total bill of materials (BOM), forming an acute hardware supply constraint with fewer than ten global suppliers capable of scaling to volume.
Drawing directly from the latest audited regulatory filings, industrial prospectuses, and factory data across major global robotics clusters, this session strips away the marketing hype to expose the stark economic and geopolitical realities of the Physical AI supply chain. Attendees will gain a granular breakdown of component-level cost structures, from strain-wave gears to linear transmission assemblies. The session will evaluate the critical "Anti-Vertical Integration" shift, analyzing how Tier-1 industrial giants are aggressively positioning to manufacture components for asset-light startups as an alternative to the vertically integrated playbooks of Tesla and Figure.
Finally, the presentation maps out the largely undisclosed cross-border dependencies currently embedded in Western robotic roadmaps, providing founders, operators, and investors with the exact frameworks required to evaluate real operational capacity, protect unit economics, and de-risk their physical scaling strategies over the next 24 months.
Humanoids TrackWhereRoom 203
-
10:30 am-11:15 amIntegrated Drive Actuator Design: Combining Gearheads, Motor, Drive and Sensors for Smarter Motion
Selecting the right motion architecture is one of the most important design decisions in robotics. This session will examine the tradeoffs between component-level and integrated actuator architectures, helping engineers understand when each approach delivers the best balance of performance, flexibility, and development efficiency.
Attendees will learn how integrated actuators can reduce engineering effort by simplifying system integration, minimizing cabling complexity, shortening testing and commissioning, and accelerating time-to-market. The session will also provide practical guidance on selecting motion solutions based on application requirements, communication protocols, performance targets, and overall system constraints, equipping attendees with a framework for making informed design decisions in next-generation robotic systems.
Technologies TrackWhereRoom 204
-
11:00 am-11:30 amSensor Solutions for Humanoids: Tools to Build the Future

Autonomous humanoid robots promise to efficiently execute a broad range of functions serving human needs – but this vision demands highly advanced sensing capabilities to perceive, navigate, and function within dynamic environments. Some sensing solutions are now ready for design-ins, while other challenges are yet to be solved.
This session explores:
-
Sensing solutions enabling today's humanoid innovations
-
Where breakthroughs are needed for exponential growth
-
The critical question: how do we give humanoids the awareness needed to operate safely alongside humans?
Engineering TheaterWhereExhibit Hall B
-
-
11:30 am-12:15 pmWhy Food is Physical AI’s Hardest Problem & Most Promising Catalyst
Physical AI has made remarkable strides, but most foundation models are trained on rigid objects. Food represents one of the most complex manipulation challenges in robotics: every ingredient is deformable, inconsistent in weight and texture, sensitive to temperature, and must be handled with calibrated force across thousands of variations.
Chef Robotics has built the largest real-world dataset of deformable material manipulation by completing over 118 million food servings in production across over a dozen food manufacturing facilities in North America and Europe. This data serves as the basis for our Food Foundation Model (FFM), which enables robots to generalize to new ingredients with minimal retraining.
This session will explore what makes food such a demanding benchmark for physical AI: the sensor-fusion challenges of grasping soft and unpredictable objects, the force-control precision required to handle fragile versus dense materials, and why real-world variation at scale is irreplaceable for training robust policies.
More importantly, we’ll show why solving food unlocks progress far beyond this industry. The techniques developed here (e.g., adaptive grasping, tactile feedback integration, high-variance training distributions) may transfer to medical devices, flexible packaging, agriculture, and other domains involving deformable materials.
Attendees will leave with a concrete framework for thinking about deformable material manipulation as the next frontier in physical AI, and evidence that real-world data at scale is what separates lab demos from deployable systems.AI TrackWhereRoom 209/210
-
11:30 am-12:15 pmStartup Boot Camp Part 2: Challenges & Best Practices for Working with Corporate Partners
Partnering with large corporations can accelerate growth, but navigating these relationships presents unique challenges. This session delivers actionable strategies for successfully collaborating with corporate partners. A startup and Industry leader share their perspectives on maximizing the value of these alliances, covering topics that include:
-
The Startup Perspective: Securing, structuring, and executing successful pilot programs and Proofs of Concept (POCs).
-
The Corporate Perspective: Inside strategies on how large enterprises evaluate, onboard, and collaborate with emerging startups.
-
Collaboration Best Practices: Bridging the gap between startup agility and corporate processes to ensure mutual success.
Business & Deployment TrackWhereRoom 207
-
-
11:30 am-12:15 pmDeployment Over Demos: Building Risk-Aware Foundation Models for Safety-Critical Environments
Despite decades of progress in robotic hardware, sensing, and motion control, the autonomous deployment and scalable operation of robots in real-world field environments remain limited. Persistent challenges arise from uncertainty, risk, and the physical constraints of dynamic, unstructured settings. In this talk, Dr. Ali Agha, CEO of FieldAI, will describe how a physics-first foundation model designed specifically for Physical AI can address long-standing autonomy challenges in robotics.
Unlike conventional vision or language models adapted for robotic systems, FieldAI’s Field Foundation Models are developed from the ground up to reason about the physical world and operate under real-world constraints. This approach enables robots to make decisions in real time and navigate dynamic, unstructured environments without relying on maps, GPS, or predefined paths. Dr. Agha will outline how this architecture-first approach to foundation model development is already being deployed and scaled successfully in industrial environments around the world.
Field Robotics TrackWhereRoom 206
-
11:30 am-12:15 pmA Joint-by-Joint Guide to Humanoid Motion
This presentation addresses the central challenge facing humanoid robot OEMs today: the gap between a working prototype and a manufacturable, scalable product. While software iterates in weeks, motion hardware moves from paper to functional sample in months and to volume production in years. That mismatch—not concept design—is the real bottleneck.
By reframing humanoid motion as a zone-by-zone architecture problem rather than a one-size-fits-all approach, the presentation outlines the unique demands of each body zone and the challenges of scaling for production. It also provides a practical framework for evaluating motion partners, highlighting comprehensive solutions and co-engineering support to simplify the path from prototype to volume production.Humanoids TrackWhereRoom 203
-
11:30 am-12:15 pmTactile Sensing Solutions for the Physical AI Era
Vision and force sensing have long been essential to robotics, but increasingly capable physical AI systems are creating new demands for robots to understand and interact with the world through touch. This session will explore how tactile sensing is evolving, where it adds value beyond vision, and how advances in sensors, data, and AI are enabling robots to handle objects and perform tasks with greater dexterity and reliability.
Technologies TrackWhereRoom 204
-
11:45 am-12:15 pmThe Robot Out of the Cage. Advanced Functional Safety for Humanoids

For fifty years, industrial robot safety has meant one thing: the cage. Put the machine behind a fence, add a light curtain, cut the power if anyone gets close. That model collapses the moment a robot leaves the cage and walks on two legs through a shared workspace.
Florian Weisshardt makes the case that humanoid robotics cannot inherit the safety playbook of industrial robotics, and shows how Synapticon's POSITRON platform is rewriting it from first principles:- Safe Torque Off does not work on two legs. A humanoid is an inverted pendulum. Cutting the power doesn't make it safe, it makes it fall. Safety for legged systems requires active safe motion, not the absence of motion.
- 50+ degrees of freedom need a new safety architecture. POSITRON delivers certified (SIL3, PLe) Safe Motion monitoring across the entire kinematic chain, including safe inverse kinematics for 7-DoF arms, a scale that traditional safety PLCs cannot reach.
- From 2D LIDAR to Safe Human Detection. Perception-based safety replaces physical guards: camera-based detection with and without AI, real-time 3D person tracking. [TODO JAN: Status SIL2/PLd bestaetigen, im Mai stand "in preparation"]
- Behavioral Safety, making black-box AI certifiable. A multimodal AI layer analyzes the robot's environment and anticipates hazards before they become incidents.
- Hardware built to the standards. POSITRON STA1 (dual 800 MHz ARM) for deterministic motion safety, POSITRON STN1 (NVIDIA IGX Jetson Thor) for the AI-heavy safety stack. Both aligned to ISO 26262, IEC 61508, ISO 13849 and ISO/IEC TS 22440, certified by TUEV Rheinland.
The takeaway: humanoids will not reach real-world deployment on the back of engineering heroics and disclaimers. They will reach it on the back of a certifiable safety architecture built for how they actually move, perceive and decide. That architecture exists today.
Engineering TheaterWhereExhibit Hall B
-
12:15 pm-1:15 pm
Networking Lunch on Exhibit Floor
Breaks/MealsWhereHall A & B
-
12:15 pm-1:15 pmWomen in Robotics Luncheon Sponsored by MassRobotics
Spend an hour connecting with robotics professionals in an environment designed for thoughtful conversation and mentorship. This lunch will include a fireside chat between Joyce Sidopoulos, Co-Founder of MassRobotics, and Bhavana Chandrashekhar, Senior Manager of Applied Science at Amazon Robotics.
This is an opportunity to form new connections, deepen relationships, and champion increased female representation in robotics leadership.
WIR LunchWhereRoom 209/210
-
12:30 pm-1:00 pmFrom First Flash to Field Repair: Servicing Robots From a Tablet

Robots can navigate warehouses, cross fields, and deliver packages autonomously. But when something breaks, how much of that autonomy survives?
This session works through the engineering questions behind keeping robots running, on the bench and in the field.
-
How do you bring up a Jetson or Raspberry Pi, and the microcontrollers behind it, the same way every time?
-
How do you ship complete software releases, firmware updates, and security patches to robots that rarely see the internet?
-
How do you keep software running with no one watching and no connection?
-
What does a technician on site need to bring a replacement board back as the same robot?
Picture a warehouse robot whose compute board dies on the night shift, a farm robot that needs a security patch in a field with no signal, or a delivery robot whose software crashes mid-route.
Three very different problems, but all exposing the same uncomfortable assumption that the right engineer, the right tools, and a reliable connection will be available when something goes wrong. That might work for a handful of robots, but it becomes a very different challenge when you're operating hundreds.
We'll show an NVIDIA Jetson going from a bare board to a running navigation stack, flashed and provisioned directly from an Android tablet without an internet connection. And when a software process crashes, we'll demonstrate how an on-device agent detects the failure and brings it back automatically, even while offline.
Along the way, we'll explore whole-system updates, remote troubleshooting, and how to make field repairs practical for technicians who weren't the ones who built the robot in the first place.
There's no shortage of conversation about what robots can do. We're interested in the work it takes to keep them doing it, long after they've gone into the field. We'll share what we've been building at DeviceNexus, what we've learned along the way, and why these challenges matter at every stage, from a fleet of one still finding its feet to operating hundreds of robots.
Engineering TheaterWhereExhibit Hall B
-
-
1:15 pm-2:00 pmFrom Models to Machines: Building the Open Infrastructure for Physical AI
AI is moving from digital environments into the physical world, where robots and autonomous systems must sense, reason, and act in real time. For robotics developers, this shift creates a new challenge: success is no longer defined by model performance alone, but by the ability to deploy AI into systems that are responsive, reliable, safe, and scalable.
This session explores the infrastructure required to move Physical AI from prototype to production in real-world robotics systems. It will examine how open, edge-first architectures can help robotics teams integrate perception, sensor fusion, motion planning, deterministic control, and real-time AI across diverse hardware and software environments.
Attendees will learn why the next phase of robotics innovation depends on more than bigger models or better simulations. It requires a practical deployment foundation that supports flexibility, interoperability, and measurable outcomes across manufacturing, logistics, field robotics, humanoids, and autonomous systems.AI TrackWhereRoom 209/210
-
1:15 pm-2:00 pmState of Robotics Investments
Where is capital flowing today, and what does it mean for robotics? Sanjay Aggarwal, Partner at F-Prime Capital, will unveil the firm's annual State of Robotics Investments report, highlighting the latest funding trends, emerging technology categories, and the sectors attracting the most investor interest. Attendees will gain a data-driven look at the current investment landscape and actionable insights into what it takes to build and finance successful robotics companies in today's market.
Business & Deployment TrackWhereRoom 207
-
1:15 pm-2:00 pmClosing the Capability Gap: A Leadership Framework for Scaling Field Robotics
Most autonomous systems work in the lab and stall in the field. An early-stage construction robot reaches a customer at roughly 80% of the capability that the customer needs. Whether that program crosses to 120% is not a hardware problem, it’s a leadership decision: where do you spend finite engineering capacity, and where do you build operator discipline instead?
Kenrick Tjandra leads robotics deployment at Raise Robotics, which holds the U.S. commercial record for autonomous robots installing building envelopes at the leading edge of buildings under construction: 10 projects, 8 states, 4,528 robot-hours, zero safety incidents. From that record, Kenrick has built a repeatable decision framework, now peer-reviewed by Project Production Institute, which answers the question every scaling robotics team faces: “do we build the product fix, or write the operator playbook?”
Kenrick will ground his talk in two failures that shaped the framework, then walk through the framework: 1) Sampling bias: your first three sites don't represent your next thirty. 2) Margin stacking: independently-defensible tolerances consume each other in series (truck → forklift → hoist → doorway → floor) until a comfortable margin is gone. 3) The framework itself: a resource taxonomy, a three-tier build-vs-workaround model, and a capability-envelope-versus-observed- variance method for anticipating where a system breaks next.
Attendees leave with a decision rule for allocating engineering capacity, a vocabulary for two hidden failure modes, and the pre-mobilization gates that keep field programs from stalling; the discipline that separates a demo from a system that scales.Field Robotics TrackWhereRoom 206
-
1:15 pm-2:00 pmBeyond the Pilot: What It Takes to Put Humanoids to Work in Manufacturing
Featuring Dr. Weiwei Chen (Founder & CEO, EmbodyX)
Humanoid robots are moving from viral demos into real factories. But getting a humanoid to run shift after shift on a live production line, matching human cycle times and hitting industrial grade requirements, takes much more than a model.
In this session, EmbodyX shares what it has learned deploying humanoids in real manufacturing environments. The talk covers the full deployment path: data collection and simulation, and running foundation models for real-time, on-premises control, and safety systems to handle unexpected simulations without putting people or equipment at risk.
Humanoids TrackWhereRoom 203
-
1:15 pm-2:00 pmHow Robotics and Physical AI are Transforming Modular Data Center Construction
Demand for data centers is surging - and building and infrastructure construction needs to keep pace. Join ABB Robotics to discover how robotic automation and emerging Physical AI technologies can accelerate data center construction, improve build quality and enhance worker safety while addressing labor shortages.
From the building structure to the equipment needed to power, distribute energy, cool and operate data centers, learn how automating repetitive, labor-intensive tasks can help teams build faster. Explore how robotics, prefabrication and modular construction can work together to streamline delivery, reduce pressure on scarce labor and scale data center capacity—and discover where automation can make the biggest difference.
Technologies TrackWhereRoom 204
-
2:00 pm-2:30 pmSimulation Said Go. The Binary Said No: Keeping Robots Inside Their Requirements in the AI Era
Featuring Stojan Gancev (VP of R&D, Aurora Labs)
Code gets written faster than anyone can read it now - humans and AI agents both - but robots still live or die by their requirements. Timing deadlines. Memory ceilings. Power budgets. This session is about reasoning straight from the architecture, from the compiled binary sitting on the target processor, rather than trusting review and simulation to catch what they structurally cannot see.
I'll show you a live demo. An innocent-looking change to a crop-spraying drone passes review and simulation but breaks its execution contract before the drone ever leaves the ground. You walk away with a pattern. It turns system requirements into enforced contracts that bite at merge time - no matter who, or what, wrote the code.Engineering TheaterWhereExhibit Hall B
-
2:15 pm-3:00 pmPhysical AI Doesn't Have a Data Problem, it Has a Deployment Problem
Featuring Rudy Cohen (Co-Founder & CEO, Inbolt)
Physical AI's consensus says progress is gated by data. More of it, better curated, better simulated. That's true in the lab. In the factory, it's the wrong constraint. After deploying vision-guided control across 100+ factories and 40 million-plus robot cycles, Inbolt CEO Rudy Cohen argues the real bottleneck is the loop between perception and motion, and the integration tax surrounding it.
This talk reframes where Physical AI actually pays off. Not in bigger models, but in closing the loop at servo frequency and collapsing the months of fixtures, wiring, and retuning that make every deployment a six-figure project. Using production data from lines at Stellantis, Toyota, and Ford, Cohen makes the case that Physical AI is already commercially proven, just not where the headlines are looking.
Key takeaways:- Where the real cost of a Physical AI deployment sits (and why the model is the cheap part).\
- Why real-time control, not model size, is the last mile.\
- A field-tested benchmark for "deployable today" vs. "impressive in a demo"
AI TrackWhereRoom 209/210
-
2:15 pm-3:00 pmThe Economics of Robotics: Is This Business Viable?
Most robotics companies don't fail because the robot doesn't work. They fail because the economics never did, or because the company ran out of cash before the economics had a chance to. Both are avoidable, and both are predictable, if you know what to look at. This session offers a set of mental models for judging whether a robotics business can create real value and capture it in a way that lasts.
We start with whether the robot is worth building at all. Most people get this wrong on both sides. They undercount the value, thinking only about labor saved while ignoring many important factors, and they underestimate the true cost. We will work through how to understand these relationships, and what raises or lowers the bar a business has to clear for success.
Then we follow the cash. How money moves through a robotics business is the clearest picture you will get of how value moves through it. Trace the cash and you can see where value is genuinely created, where it is captured, and where it quietly leaks away before anyone gets paid. We will look at why robotics has one of the most punishing cash cycles in hardware, why the most popular business model makes it worse, and how the best operators turn that flow into a durable advantage. We will conclude by discussing how companies can think creatively about raising capital.
You will leave with a clearer way to read any robotics business.Business & Deployment TrackWhereRoom 207
-
2:15 pm-3:00 pmLessons Learned Building Reliable Outdoor Robots
As robots move beyond controlled indoor environments and into agriculture, mining, construction, defense, ports, and critical infrastructure, the engineering challenges become significantly more demanding. Weather, dust, moisture, vibration, uneven terrain, and limited access to maintenance introduce failure modes rarely encountered during lab testing.
Drawing on years of experience supporting autonomous systems in some of the world's harshest environments, this panel will share practical lessons for designing, building, and deploying robots that can withstand real-world field conditions. Attendees will learn best practices for environmental hardening, power architecture, autonomous operations, charging infrastructure, reliability engineering, and long-term fleet maintenance.
Through real-world deployment examples, the session will demonstrate how engineering decisions impact reliability, uptime, operational efficiency, and total cost of ownership. Engineers will leave with practical strategies for building resilient robotic systems that can operate reliably despite exposure to rain, dust, extreme temperatures, vibration, and other environmental challenges.
Field Robotics TrackWhereRoom 206
-
2:15 pm-3:00 pmA Crash Course in Force & Torque Sensing for Humanoids
Featuring Robert Brooks (Founder & CEO, ForceN)
Learn about the many ways force & torque sensors can be used in humanoids and how to avoid breaking them while doing it. We'll summarize learnings from working with many of world's top humanoid companies over the past three years and best practices we can incorporate from over a 10 years of force/torque sensor development in surgical, logistics, and field robotics field with the goal of accelerating the production & field readiness of humanoids.
Humanoids TrackWhereRoom 203
-
2:15 pm-3:00 pmFunctional Safety for Humanoid Actuators
The future of humanoid robotics depends on trust. That trust begins with safe, reliable motion. Humanoids are expected to work safely alongside people, making actuator design a critical element of functional safety. Novanta's Jordan Schaeffer explores how safe motion, torque control, fault detection, and actuator-level safety functions can help reduce risk while maintaining performance. Learn about the challenges of human-robot interaction and the motion technologies enabling the next generation of safe, collaborative humanoid systems.
Technologies TrackWhereRoom 204
-
3:00 pm-3:30 pm
PM Networking Break
Breaks/MealsWhereHall A & B
-
3:30 pm-4:15 pmHow Agentic Coding Empowers Lean Teams to Build Faster
Building a robot from scratch has traditionally required massive engineering teams and years of sluggish development cycles, but agentic coding is fundamentally upending this paradigm. In this session, Anthony Jules from Robust.AI will explore how AI agents are transforming the robotics development lifecycle right from the initial design phase. By integrating agent-driven workflows from day one, engineers can now automate complex coding tasks, rapidly iterate on early prototypes, and overcome technical hurdles at unprecedented speeds.
AI TrackWhereRoom 209/210
-
3:30 pm-4:15 pmThe RaaS Playbook: Pricing, Service & Scale
Robots-as-a-Service (RaaS) has become a popular path to market for commercial robotics, but building a successful RaaS business requires far more than just offering a subscription model. From pricing and financing to fleet management, customer support, and service-level agreements, companies must balance recurring revenue with the operational realities of deploying robots in the field.
In this session, experienced robotics leaders will share best practices for launching, operating, and scaling a successful RaaS business. Learn how to structure contracts, price services, define customer success metrics, manage maintenance and uptime, and determine when RaaS is the right fit—or when a traditional capital purchase model makes more sense. Attendees will leave with practical strategies for building sustainable RaaS offerings that deliver value for both robotics providers and their customers.
Business & Deployment TrackWhereRoom 207
-
3:30 pm-4:00 pmShared Spaces: Navigating Challenges for Indoor Robots
Consumer-facing spaces are dynamic, crowded, and highly unpredictable, requiring robots to safely navigate around constantly changing obstacles, operate in diverse lighting conditions, navigate narrow passages and cluttered spaces, integrate with existing infrastructure, and balance autonomy with safety. As robots move beyond warehouses and other controlled environments and into everyday life, solving these navigation challenges will be critical to achieving broad consumer acceptance and large-scale deployment.
Engineering TheaterWhereExhibit Hall B
-
3:30 pm-4:15 pmDesigning Dexterous Learning Robots for Outdoor Labor
Featuring Brent Shedd (Founder & CEO, AgRobotics)
High-value outdoor labor is running out of hands. Across agriculture, construction, landscaping, and the skilled trades, skyrocketing costs, tightening regulation, and collapsing workforce availability are creating compounding pressure on industries that cannot move indoors and cannot wait. Conventional automation hasn't answered the call. These environments are unstructured, unpredictable, and physically demanding in ways that have stymied robotics for decades.
This session introduces a new category of dexterous, learning field robots built specifically for outdoor labor - not adapted from warehouse or lab platforms, but engineered from the ground up for adverse, uncontrolled environments. The $1.8B U.S. strawberry harvest serves as the proving ground: if a robot can handle soft-body, deformable objects with precision across thousands of acres of changing terrain, the underlying architecture transfers across countless outdoor labor uses.
Attendees will meet DEX-26, a purpose-built dexterous robot designed to work directly in the furrow - and eventually on the jobsite, in the landscape, and across the trades. The session examines the VLA model architecture making this possible, and the real-world data methods and strategies required to train dexterous manipulation at scale outdoors.
Drawing on thousands of hours of prototyping, field testing and business building, this talk delivers a practitioner-level framework: what's working, what's hard, and what a realistic path to commercial deployment actually looks like.Field Robotics TrackWhereRoom 206
-
3:30 pm-4:15 pmCan Humanoids be Safe?
As AI becomes increasingly central to the development of humanoids, the demand for rigorous safety frameworks grows more urgent. This presentation examines how traditional functional safety principles, particularly those outlined in IEC 61508, can provide a foundation for the safe integration of AI in humanoids, while highlighting where conventional methods fall short. Drawing from deep expertise in functional safety, this talk explores the intersection of time-tested safety practices and the unique demands of AI-driven humanoids. We’ll assess how core concepts like Safety Integrity Levels (SILs), hazard analysis, and systematic failure handling apply to human-robot interaction, adaptive learning, and unpredictable environments. Attendees will gain insight into both the continuity and the evolution of safety thinking as it adapts to intelligent, human-like machines operating in dynamic, real-world contexts.
Humanoids TrackWhereRoom 203
-
3:30 pm-4:15 pmWhat AI-Defined Automation Looks Like in Production
The PLC made automation programmable. AI is making it more adaptive and accessible. Vention CTO Francois Giguere explains how two layers of intelligence work together: Physical AI at the machine, handling perception, motion planning, and variability, and Agentic AI across the system, from cell design and programming to fleet monitoring. Using real deployments with leading manufacturers, he breaks down what scales, what it costs, and where conventional automation still wins. Attendees leave with a framework for evaluating AI-defined automation in their own operations.
Technologies TrackWhereRoom 204
-
4:30 pm-5:15 pmBeyond the Demo: AI in Production Robotics
Featuring Eugene Demaitre (Editorial Director, The Robot Report), Bhavana Chandrashekhar (Senior Manager of Applied Science, Amazon Robotics), James Davidson (Chief AI Officer, Teradyne Robotics), Ali Raja (Vice President, Data Centers, ABB Robotics), Ram Devarajulu (VP of Deep Tech Innovations, Cambridge Consultants), Jon Battles (Vice President of Technology Strategy, Cobot)
Everyone is talking about AI. This panel is about what happens after the demo. Hear from leaders at Amazon Robotics, Teradyne Robotics, ABB Robotics and Cobot as they discuss how they're applying AI to robots operating in real customer environments. From perception and autonomy to fleet optimization, maintenance, and software updates, learn what it takes to build AI that is reliable, scalable, and delivers measurable business value.
GeneralWhereTheatre - Floor 2
-
5:00 pm-6:30 pm
Networking Reception Sponsored by Harmonic Drive
Exhibitor HallWhereHall A & B
-
8:00 am-3:00 pm
Registration Hours
RegistrationWhereMain Lobby Foyer
-
9:30 am-10:15 amState of Humanoids
Humanoid robots have captured the world's attention, attracting billions in investment and generating excitement about the future. But beyond the viral videos and ambitious promises, where does the technology really stand?
This keynote brings together industry leaders to examine the current state of humanoids through the lens of real-world deployment. The discussion will explore the capabilities humanoids are delivering today, the technical and operational challenges that remain, and the lessons emerging from early industrial implementations. Panelists will also address key topics including safety, standards, economics, and the milestones that must be reached before humanoids can achieve widespread adoption across manufacturing, warehousing, and logistics.
Attendees will leave with a clear, practical understanding of where humanoids create value today and what it will take for the technology to scale.
GeneralWhereTheatre - Floor 2
-
10:00 am-3:00 pm
Exhibitor Hall Hours
Exhibitor HallWhereHall A & B
-
10:15 am-10:45 am
AM Networking Break
Breaks/MealsWhereHall A & B
-
10:45 am-11:30 amEgocentric Robot Learning: Teaching Robots Through Human Experience
Featuring Ted Larson (Co-Founder @ CEO, OLogic)
Egocentric robot learning is emerging as a powerful approach for developing more capable, adaptable, and scalable robotic systems. By learning from first-person human demonstrations captured through wearable cameras, teleoperation systems, and embodied AI platforms, robots can acquire complex manipulation and decision-making skills in ways that more closely mirror human learning. This paradigm has gained significant momentum through advances from organizations such as NVIDIA, Google DeepMind, Meta, and a growing ecosystem of robotics startups focused on foundation models, imitation learning, and embodied intelligence.
As labor shortages, increasing customization demands, and operational complexity continue to challenge industries, egocentric learning offers a path to reducing the time, cost, and expertise traditionally required to deploy robotic systems. By enabling robots to learn directly from human behavior, organizations can accelerate automation across manufacturing, logistics, healthcare, service robotics, and other commercial applications.
In this session, Ted Larson, CEO of OLogic, will explore the foundations of egocentric robot learning, the technical challenges involved in building these systems, and their implications for the future of commercial robotics.
Attendees will gain insight into how advances in embodied AI, multimodal data collection, and large-scale robot learning are reshaping robot development and deployment. The presentation will examine both the opportunities and challenges ahead, highlighting how egocentric learning can help bridge the gap between research innovation and scalable, real-world robotic applications.AI TrackWhereRoom 203
-
10:45 am-11:30 amChina’s Physical AI Ecosystem in 2026: Competitor, Supplier, Benchmark
The relationship with China is complex. While powerful voices call for de-risking and the U.S. is working on a national robotics strategy, many startups (even some who claim the opposite) are sourcing heavily from China. Western companies claim leadership in robot foundation models, but does this hold true? The landscape of emerging Chinese robotics companies is highly uneven, yet we see promising approaches in robot brains and data flywheels there, too. In this talk, I will share the latest insights from our work on the ground in China and put them into global perspective. While some of our clients use the Chinese ecosystem for their global success, for others it is only a benchmark — but an important one.
Business & Deployment TrackWhereRoom 207
-
10:45 am-11:30 amAn Introduction to Robotic Tools for Overhead Power Line Operations
This presentation explores how drone-deployed robotic tools can reduce worker exposure to hazards and improve the efficiency of overhead power line operations. Using FulcrumAir’s technology as a practical example, participants will learn how robotics support spacer, bird diverter, and marker ball installation, as well as lead line pulling.
The session examines mobilization, installation precision, live-line capabilities and opportunities to complete work without helicopters or required outages.
Field Robotics TrackWhereRoom 209/210
-
10:45 am-11:30 amPrecision, Power, Perception: Enabling Next-Gen Humanoid Actuation
As humanoids grow in complexity, the demands on their actuation systems escalate, requiring unprecedented levels of precision, power density, and energy efficiency. This session will dissect the critical technological advancements that are making next-generation robotic actuation possible. We will explore three solutions that address these challenges:
-High fidelity tunnel magnetoresistance (TMR) sensors for motor actuator torque and position control for greater accuracy and speed in compact form factors.
-The strategic deployment of high-voltage GaN-based power devices to improve power density and reduce I²R losses in robotic joints.
-The integration of embedded AI at the sensor edge to perform pattern abstraction and information representation while minimizing data movement and reducing power consumption.
Drawing from real world examples and the latest research, attendees will discover how these integrated technologies converge to create more capable, efficient, and intelligent robotic systems.Humanoids TrackWhereRoom 206
-
10:45 am-11:30 amWhy Physical AI Fails in the Real World
A model that performs well in simulation is an important milestone. But it is not proof that a robot is ready for the real world.
The hardest challenges in Physical AI emerge when learned behaviors encounter changing environments, imperfect sensors, limited compute, and unexpected edge cases. This panel will explore what it takes to bridge the gap between training and deployment, including the engineering disciplines required to build reliable robotic systems.
Experts will discuss how teams are addressing challenges across perception, decision-making, and manipulation to create robots that can operate safely and predictably outside the lab.
Technologies TrackWhereRoom 204
-
11:45 am-12:30 pmBuilding Physical AI that Scales
Recent advances in foundation models, vision-language-action systems, and edge AI are expanding what robots can sense, reason about, and do. The next challenge is turning those capabilities into autonomous systems people can trust in the real world, whether restoring mobility through intelligent prosthetics, delivering critical medical supplies, or helping workers operate more safely and productively alongside intelligent machines.
As robots become more capable, engineering priorities increasingly shift toward system-level design. Success will depend on how intelligence is distributed across the robot, how learned and deterministic behaviors interact and how compute platforms balance real-time responsiveness, efficiency, and safety.
Drawing on lessons from across the robotics ecosystem, this session will examine key compute principles behind trusted physical AI, including how intelligence is distributed across cloud and edge compute, the coordination of learned and deterministic behaviors, and how developers can design systems that remain adaptable, reliable, and safe as they scale to production.
Attendees will leave with a practical framework for building trusted physical AI systems that extend human capability while meeting the real-world demands of safety, efficiency, and scale.AI TrackWhereRoom 203
-
11:45 am-12:30 pmReshoring With Robots: A Policy Discussion
Featuring Ellie Sakhaee (AI Policy Manager, Google), Eugene Demaitre (Editorial Director, The Robot Report), Dr. Henrik I Christensen (Distinguished professor of computer science, UC San Diego), Brendan Schulman (VP of Policy & Government Relations, Boston Dynamics), Michael Bearman (Chief legal and safety officer, Vecna Robotics)
Robotics is a key enabler of reshoring production to the U.S., but there is little consensus on how to best encourage it. Most industrial automation is made by foreign-owned companies, but can recent policies such as tariffs and FCC import restrictions help build back manufacturing? This expert panel, which includes representatives of leading robot manufacturers and industry analysts, will examine the need for a national robotics strategy and debate industrial priorities.
Business & Deployment TrackWhereRoom 207
-
11:45 am-12:30 pmSparkE-B: Building a Robot for Safer Electrical Work
Electrical circuit energization is one of the most hazardous tasks on a construction site. During initial power-up or post-maintenance, arc flash events can generate extreme temperatures and expose electricians to potentially catastrophic risks—even with extensive protective equipment.
In this session, Hardhat Robotics will introduce SparkE-B, a mobile robot designed to remove workers from the arc flash zone by enabling electricians to remotely perform circuit energization. The session will explore the challenges of deploying robots in dynamic construction environments, the engineering behind safety-critical robotic systems, and how AI-enabled autonomy could help skilled tradespeople dramatically increase productivity.
Field Robotics TrackWhereRoom 209/210
-
11:45 am-12:30 pmThe Dexterous Hand: Delivering Force and Signal via Synthetic Tendons
As humanoid hands move from research laboratories into commercial deployment, industry discussions often focus on AI, perception, dexterity, and actuation. Yet one of the most critical enabling technologies remains largely overlooked: the force and signal transmission system that connect actuators to human-like motion.
This presentation explores the emerging role of synthetic tendons, Bowden cable architectures, and integrated sensing systems in enabling lightweight, efficient, and scalable dexterous hand designs. Attendees will gain insight into how tendon-driven systems reduce mass, improve compliance and safety, and create new opportunities for dexterous manipulation. The session will examine key engineering challenges including fatigue life, creep, abrasion resistance, friction management, routing constraints, and reliability under millions of flex cycles.
Drawing on real-world development and manufacturing experience, the presentation will also discuss how advanced materials, hybrid constructions, embedded conductors, and sensing technologies are creating intelligent tendon systems capable of transmitting force, power, and information simultaneously. Finally, the talk will address one of the industry's most pressing challenges: transitioning from promising prototypes to production-ready humanoid platforms. Participants will learn how scalable manufacturing, validation testing, quality systems, and supply-chain readiness can accelerate commercialization and improve long-term robot reliability.
Humanoids TrackWhereRoom 206
-
11:45 am-12:30 pmHuman-Robot Interaction: Bridging Academia and Industry
Human-robot interaction (HRI) is one of the biggest unlocks left in industrial automation to enable stronger safety, but most HRI research happens on university campuses that produces data that rarely survives contact with a live production floor. Running studies inside a working manufacturing site is logistically hard enough that most academia–industry partnerships never get past the first meeting. Faraz Yousuf (Eli Lilly) and Dr. Anastasia Kouvaras Ostrowski (Purdue) share what they learned running HRI field studies with manufacturing and warehouse operators around live AGVs, some early findings of people's perception of robots on the manufacturing floor and the playbook they built to make the partnership repeatable.
Technologies TrackWhereRoom 204
-
12:30 pm-2:00 pm
Networking Lunch
Breaks/MealsWhereHall A & B
-
2:00 pm-2:45 pmReal-Time Robot Troubleshooting with AI Agents and MCP
When a deployed robot fails, the investigation usually lands on an experienced field engineer: pulling logs, echoing topics, and correlating sensor data with commands. It is slow, it does not scale with the fleet, and the knowledge rarely leaves that engineer's head. This talk shows how the right open-source tooling lets AI agents take the first troubleshooting pass directly on a deployed robot.
After a quick introduction to the Model Context Protocol (MCP), we show examples on a real industrial robot deployed in a warehouse: an agent tracing a fault from symptom to root cause. We then look at the benefits of adding context such as technician manuals or the codebase, which extends the same approach from field troubleshooting to developer velocity. We also touch on using the agent to turn a one-off investigation into reusable inspection scripts.
We close with how this connects to the Open Source Robotics Alliance's Physical AI Initiatives, and how to run it on your own robots today.
AI TrackWhereRoom 206
-
2:00 pm-2:45 pmThe Best Models Still Reason Visually Like Toddlers
Frontier AI models score 80–90% on standard benchmarks like RKGI, yet when tested on visual tasks any 3-year-old handles effortlessly (like counting objects in an image), those same models fall to pieces. Andrew watched this gap widen firsthand during his 14 years at Google Brain and DeepMind, where he co-led development on GLaM, PaLM 2, and Gemini. The problem is that most models hit high RKGI scores not through genuine visual understanding, but by coding – a workaround that scores well and reveals little. Strip that away and you're left with systems that struggle to solve a simple crossword puzzle, identify what's the same or different across two images, or navigate a basic 3D view. These tasks are essential to achieve human-level reasoning capability.
Andrew will dig into why the current eval landscape systematically overstates capability, the structural reasons it does so, and how we got here from the viewpoint of someone who was inside a leading frontier lab. He will close with what he believes a more rigorous, consensus-driven eval framework needs to look like, and why the field needs to build one before the next generation of visual systems ships into the real world. Fixing visual reasoning starts with fixing how we measure it. For engineers building on top of these models today, whether that's document understanding, robotic perception, medical imaging, or any system where visual perception context matters, the cost of getting this wrong is already showing up in production.AI TrackWhereRoom 203
-
2:00 pm-2:45 pmThe ROI of Reality: Investing in the Physical AI Frontier
The generative AI boom captivated the market, but software eating the world was only phase one. The next massive value unlock is Physical AI, where intelligence safely interacts with the real world through robotics, mobility, and autonomous agents.
While large language models have mastered the digital realm, translating cognitive capability into physical action introduces profound complexities. Unlike pure software, physical systems must contend with gravity, spatial reasoning, and chaotic human environments. These friction points present formidable challenges, but they also create deep competitive moats for the founders who solve them through seamless hardware-software convergence.
Toyota Ventures recognized this structural market shift early, having backed companies working on these foundational building blocks of physical AI and automation since the firm’s founding in 2017.
In this session, David Sokolic explores the transition from static digital outputs to real-world action. Drawing on deal-flow insights and lessons from Toyota Ventures’ pioneering deep-tech portfolio, David will break down what it takes to build, fund, and scale intelligent physical systems. He will highlight the most lucrative white-space opportunities for founders, and reveal why mastering real-world unpredictability is the key to capturing AI's next great market opportunity.Business & Deployment TrackWhereRoom 207
-
2:00 pm-2:45 pmHow to Pick Winners in Robotics: What Programs and IP to Pursue and Why
Robotics does not exist in a vacuum. The value robotics can provide is derived from its application to real world problems. A lot of robotics ideas are technically sound but lacking in business validation. A lot of contract robotics engagements are destined for failure. This talk aims to go through some of the methodologies that I use when evaluating the viability of a business case for robotics and also whether or not to accept a partner engagement. I will give real world examples of projects we turned away and startup ideas I decided not to pursue and why as well as programs that were successful and why those made sense to pursue at the time. Specifically, this talk aims to cover:
- Knowing when to walk away
- Partner assessment
- Evaluating business cases
- Evaluating mission alignment
- Evaluating viability for IP development
Technologies TrackWhereRoom 209/210
-
2:00 pm-2:45 pmIntelligence Behind Warehouse Robotics: Scaling Operations with AI & Software
Deploying autonomous mobile robots is only the first step in warehouse automation. The real challenge is scaling those systems across growing operations while maintaining productivity, flexibility, and operational control.
In this session, Jingwei Tan, System Consulting at Geek+, will explore how AI-powered software transforms fleets of autonomous mobile robots into an intelligent, coordinated warehouse system. Attendees will learn how Geek+ combines advanced robotics with intelligent fleet management, task orchestration, traffic optimization, and real-time decision-making to maximize throughput and adapt to changing business demands.
The presentation will examine how software, not just hardware, enables organizations to expand from a single automated workflow to enterprise-scale warehouse operations, while reducing complexity, improving resource utilization, and accelerating ROI. Through practical deployment examples, attendees will gain insights into the key architectural principles required to successfully scale warehouse automation using AI.
Technologies TrackWhereRoom 204
-
3:00 pm-3:45 pmClosing Day 2 Keynote: Morgan Pope on creating robotic characters
GeneralWhereTheatre - Floor 2