# Robots Scaled Mount Tai Stairs with Cargo in Field Trials

- Published: 2026-11-07 05:31 UTC
- Section: Machines & Making › Robots
- Tags: quadruped robot stairs, humanoid balance, Mount Tai robotics, Unitree robot trial, cargo legged robot
- URL: https://realtech.media/news/rt-8622e28d47312e8c3c50efdc_robots_scaled_mount/

**Legged systems and exoskeletons competed on rugged stone staircases in China, testing balance and mountain transport.**

![Robots Scaled Mount Tai Stairs with Cargo in Field Trials](https://realtech.media/media/2026/11/4192cdc241283476.jpg)
*chip.pl*

On November 2, nearly 90 teams testing humanoid platforms, quadrupedal machines, and mobility exoskeletons gathered at Mount Tai in Shandong Province for the World Mountain Robot Challenge, as reported by chip.pl. The main route stretched from Hongmen to Zhongtianmen over 3.9 km, covering more than 3100 stone steps with an elevation gain of approximately 600 meters [1]. The event tested machine locomotion on irregular natural paths rather than structured laboratory surfaces.

Organizers divided the challenges into speed trials and freight runs. In the logistics categories, machines navigated a shorter track of roughly 3.3 km with over 2100 steps while carrying delicate loads such as watermelons and eggs, with some payloads reaching approximately 13 kg [1]. The quadruped developed by Beijing Lingyi Technology, part of the Unitree group, won the main speed run in 43 minutes and 41 seconds [1]. The fastest humanoid from the Zhishen Lingyue team completed a significantly shorter segment in 27 minutes and 26 seconds, while the Lingshi team from Linxai Intelligence secured the cargo category with a time of 1 hour, 51 minutes, and 29 seconds [1]. Field hazards caused several mechanical failures, including one humanoid falling on cobblestones and another tumbling down a staircase.

## Sensor Suites and Navigation Strategies

Participating groups approached path planning and state estimation through distinct engineering methods. Engineers from the Uniubi team utilized light detection and ranging sensors to emit laser beams, generating three-dimensional maps of the surrounding terrain to detect step profiles, loose stones, and physical barriers before stepping [1].

Conversely, the Genisom group used a blind locomotion setup that bypassed environmental imaging entirely. This system relied strictly on ground contact signals from foot sensors, joint load torque, and torso orientation telemetry, dynamically updating stride dimensions to maintain system equilibrium [1].

## Automated Logistics on Mountain Routes

Real-world trials at Mount Tai build on past logistical tests, including a 2026 deployment where a quadruped transported mountain water uphill and descended with refuse [1]. That utility system handled payloads ranging between 30 and 45 kg, operated for 4 to 6 hours on a single battery pack, and relied on vision cameras paired with range sensors [1].

Transporting fragile cargo on extreme inclines requires motion controllers that damp accelerations to preserve payload integrity while ascending stone steps. Industrial adoption along unpaved paths depends on extended mean time between failures, power draw, and maintenance overhead rather than isolated ascent speeds.

## Sources

1. [chip.pl](https://www.chip.pl/2026/11/chiny-zawody-robotow-gora-tai)

---

Source: RealTech, realtech.media — https://realtech.media/news/rt-8622e28d47312e8c3c50efdc_robots_scaled_mount/