Warehouses Just Got a Robot That Can Unload Trucks Like a Human — FedEx Is Already Using It

Warehouses Just Got a Robot That Can Unload Trucks Like a Human — FedEx Is Already Using It

7 min read•Jun 13, 2026•
Anna Kowalski
Anna Kowalski

Dexterity's new Mech robot can unload trailers with two arms and human-like dexterity, tackling one of the most injury-prone jobs in logistics. FedEx has already deployed it to free workers from dangerous, repetitive lifting while keeping operations running.

The Brutal Problem of Truck Unloading

Unloading a 53-foot trailer is one of the most physically punishing jobs in logistics — and one of the hardest to automate. Workers expend enormous energy lifting, twisting, and carrying boxes that can weigh 50–70 lbs each, often in non-climate-controlled trailers during summer heat. The Bureau of Labor Statistics reports that material movers suffer over 100,000 musculoskeletal injuries annually in the U.S. alone, a leading cause of warehouse worker turnover, which runs above 40% in many distribution centers.

Despite a decade of warehouse automation investments, truck unloading remained stubbornly analog. Fixed conveyor systems can't handle the chaotic, irregular mix of boxes inside a trailer. Single-arm robotic cells require precise box placement unrealistic in real-world freight. Humanoid general-purpose robots remain years away from reliable commercial deployment. The result: distribution centers have spent millions on automation for sorting and packing, but the back door — the trailer dock — remains labor-intensive and dangerous.

Enter the Mech: A Robot Built for the Worst Warehouse Job

Dexterity, a Redwood City-based startup, created the Mech as the first robot purpose-built to unload trucks with human-level dexterity. The machine rides on a mobile base and uses two independent robotic arms, each equipped with tactile sensors (pressure-sensitive "skin" that detects grip force in real time). A computer vision system scans the trailer interior, identifies box sizes and orientations, and plans a sequence of picks that avoids collisions between the two arms and the surrounding trailer walls.

The robot is not a humanoid — it doesn't walk or mimic human form — but its dual-arm design is optimized for the single most common intralogistics task: moving irregular boxes from an unstructured environment (the trailer) into a structured one (a conveyor or pallet). Dexterity calls it a "superhumanoid" because it exceeds human capabilities in reach, endurance, and lifting strength while matching human dexterity. It can work continuously for hours without breaks, doesn't get heat-stressed, and can be reprogrammed for different box types on the fly.

FeatureThe MechTypical Human WorkerFixed Automation
ArmsTwo autonomous, coordinatedTwo arms, limited reachZero or one, fixed position
Box handlingIrregular shapes and sizesAny, but fatigue-limitedUniform boxes only
Continuous operation24/7 (with charging breaks)8-10 hours with breaks24/7 but limited to structured input
Injury riskNoneHigh (MSDs, strains)Low (single-purpose)

How the Mech Works: Two Arms, Sense of Touch, and AI Planning

Dexterity's robot combines three technical layers that make truck unloading feasible: tactile sensing, AI planning, and real-time control via EtherCAT. The tactile sensors on each gripper — essentially force-sensitive pads — allow the robot to adjust grip strength instantly. A box that starts to slip is caught, not crushed, and the robot adapts its strategy based on feedback from each item.

The physical AI software (Dexterity's term for AI that understands the physical world, not just text or images) uses camera feeds to build a 3D model of the trailer space, then plans a picking sequence that coordinates both arms to avoid collisions. This is not a two-armed system where one arm waits for the other; both work simultaneously, unloading boxes at a rate comparable to a trained human worker — about 600–800 boxes per hour in early deployments.

Beckhoff Automation logo

The robot runs on a Beckhoff unified control platform using EtherCAT, an industrial Ethernet protocol that ensures microsecond-level synchronization between the arms, vision system, and mobile base. This allows the Mech to react to changing conditions — a box that shifts while being picked, or a suddenly rearranged stack — without pausing the operation. Integrated diagnostics and safety features let warehouse operators commission new units quickly and troubleshoot remotely, essential for scaling a fleet across multiple docks.

FedEx as Proof of Concept

Delivery giant FedEx has already collaborated with Dexterity to optimize trailer loading and unloading at select facilities. While specific deployment details remain under wraps, FedEx's involvement signals a serious vote of confidence. The company moves roughly 15 million packages per day and operates one of the largest private fleets of trucks in the world. Any automation that can reduce dock injuries and speed up turnover of trailers (idle trailers cost carriers $100–$200 per hour in detention fees) directly improves the bottom line.

Dexterity's approach is distinct from other truck-unloading robots like Pickle Robot (which uses a single arm and conveyor) or Boston Dynamics' Stretch (also single-arm, primarily for unloading). The Mech's two arms allow it to handle the chaotic box stacks common in actual freight — where boxes are jammed together, stacked at angles, and mixed with dunnage (packing materials). The dual-arm design also opens the possibility of palletizing incoming goods directly, not just dumping them onto a conveyor.

What This Means for Buyers

For distribution center operators, the Mech represents the first commercially viable solution to the truck-unloading labor shortage. Here's what to consider before buying:

  • ROI timeline: At an expected price point of $250,000–$400,000 per unit (based on comparable dual-arm industrial robots), replacing one worker per shift at $45,000/year fully loaded yields a payback of 2–3 years. If the robot runs three shifts, payback drops under one year.
  • Deployment requirements: The mobile base means no permanent infrastructure — it rolls into any standard dock door. But the facility needs adequate floor space for the robot to maneuver and a charging station. EtherCAT networking may require a dedicated industrial switch.
  • Service and support: As a startup, Dexterity's support network is narrower than established players like Fanuc or ABB. Operators should negotiate service-level agreements upfront.
  • Comparison to alternatives:
SolutionCost per unitThroughputBox varietyTraining neededInstallation time
Mechanized unloader (e.g., extendable conveyor)$100K–$200KHigh for uniform freightLowLowDays
Single-arm unloading robot$150K–$250KModerateMediumMediumWeeks
Dexterity Mech$250K–$400KHighHighMediumWeeks
Human labor (annual per dock position)$45K–$55KVariableHighLowHours

Buyers should also consider fleet scalability: Dexterity claims the Beckhoff control platform enables easy fleet management — one operator can monitor several Mechs from a single dashboard. That's critical for facilities with 10+ dock doors.

Conclusion

Dexterity's Mech robot tackles the hardest problem in warehouse automation: unloading trucks full of chaotic mixed freight, a task that injures tens of thousands of workers each year and bleeds billions in turnover costs. With FedEx already testing the system and a scalable control architecture from Beckhoff, the Mech bridges the gap between manual labor and full automation. For distribution center operators squeezed between rising e-commerce volumes and shrinking labor pools, the era of robot truck unloaders has quietly begun.

Arizona appeals court vacates manslaughter sentence after AI video

An Arizona appeals court vacated the 10.5-year sentence of Gabriel Horcasitas while upholding his manslaughter conviction, first reported by Nytimes. The case returns to Maricopa County Superior Court for resentencing without the video, after judges found that it presented scripted statements as if the victim himself were speaking in court.

The three-judge panel said the video generated a likeness of Christopher Pelkey’s voice and appearance but did not reflect actual events. It found that allowing and relying on the video made the sentencing fundamentally unfair, and noted that no prior Arizona case had addressed the admissibility of such a depiction at sentencing.

The judges said a victim’s right to speak cannot override a defendant’s right to be sentenced on accurate, reliable information. They said the video collapsed the distinction between the family’s belief about what Pelkey would have said and Pelkey’s own voice and opinions.

The ruling distinguishes family members speaking about Pelkey from a generated likeness that appeared to speak for him.

Pelkey’s sister, Stacey Wales, presented the video during Horcasitas’s sentencing alongside victim-impact statements from family and friends. Wales wrote the script and said her husband and the couple’s longtime business partner helped create the video using Pelkey’s voice from a YouTube video and his face and torso from a funeral-service poster.

Judge Todd F. Lang praised the video as genuine, then imposed the maximum sentence of 10.5 years, more than the nine years prosecutors had sought.

Wales said nobody intended to make the court believe Pelkey was alive or that he had recorded the video before his death. She said she disagreed with the ruling and argued that families use slide shows, collages, hypothetical conversations and poetry to convey grief.

Wales compared the AI video with photography, saying it took 15 years of landmark cases around the 1860s before photography was widely accepted in courts.

The case returns to Maricopa County Superior Court for a new sentencing hearing without the AI-generated video.