Frankly speaking, we've stumbled through quite a few pitfalls with our cleaning robot navigation projects. From early 2D radar solutions to now upgrading to M360, it's been like feeling our way in the dark. Our old robots used to wander around shopping malls like drunkards, frequently bumping into shelves and completely shutting down when lights went out at night. After switching to M360, the situation finally took a real turn for the better.
Real Pain Points in Cleaning Scenarios
Shopping mall shelf arrangements are basically a nightmare for robots. Those shelves all look identical, and 2D radars simply can't tell which row they're in. The result is robots spinning in place or bumping into obstacles. We tested this at a large supermarket in East China, and the robot kept going back and forth in the same area with cleaning efficiency that was terribly poor.
Things get even more problematic in office buildings at night when lights are turned off. 2D radars depend on ambient light and completely lose their navigation capabilities once it's dark. Cleaning staff have to follow the robots around turning on lights, and the electricity bill for a month would be enough to buy two new radars.
How M360 Solves These Problems
5cm Near Blind Zone is Essential
M360 can detect obstacles as close as 5cm, which is double the improvement over MID-360's 10cm. This might seem insignificant, but it's crucial in cleaning scenarios.
Imagine cleaning robots navigating through shelf gaps. Those 5cm-high steps, reflective markers on the floor - MID-360 can't see them at all, but M360 can easily identify them. During factory testing, we found that this 5cm difference directly prevented robots from getting stuck at shelf bases.
70° Vertical FOV Advantages
With a 70° vertical field of view, M360 has 11° more than MID-360. What does that mean? MID-360 can only see the ground and ceiling in front of the robot in separate parts, while M360 can simultaneously handle both ceiling shelves and ground details.
In shopping malls, this advantage is particularly significant. M360 can not only see the point clouds of shelves themselves, but also simultaneously acquire ceiling and shelf top data, forming a three-dimensional spatial positioning. This way, robots don't get lost when working in the same area due to similar features.
30% Lower Power Consumption is Practical
M360's power consumption of 4.5W is 30% lower than MID-360's 6.5W. Don't underestimate this 1.5W difference - for a cleaning robot working 8 hours a day, the electricity saved in a year is enough to buy two new radars.
Lower power consumption means longer battery life. Previously, robots using MID-360 needed two charges to complete cleaning in an office building. Now with M360, one charge is sufficient. Cleaning staff don't need to swap batteries midway, and efficiency improves significantly.
Dual Echo Penetration Capability
Frankly speaking, rainy weather is when radar failure is most feared. Earlier robots using MID-360 often lost targets in rainy and foggy conditions, or produced大量噪点. M360's dual echo mode can penetrate raindrops, maintaining stable operation even in harsh weather.
Once we tested this in an underground parking lot, M360 could see through ground water fog to detect obstacles, while MID-360 was basically ineffective. This is crucial for cleaning robots because cleaning tasks can't be interrupted due to weather.
Actual Performance in Different Scenarios
Shopping Mall Environment
What's most headache-inducing in shopping malls are the shelf arrangement areas. Those shelves all look identical, and traditional 2D radars simply can't distinguish for positioning. M360 achieves fast positioning through ceiling features, allowing free navigation between shelves.
We tested at a large supermarket in East China, where the M360 LiDAR could simultaneously recognize shelf, ceiling, and ground feature points, forming stable 3D positioning. Robots no longer wander back and forth in the same area, and cleaning efficiency improved by 40%.
Office Building Environment
Office building lobbies have heavy human traffic and frequent floor cleaning. M360 is unaffected by light conditions and can work independently even when lights are off at night. We tested at an office building in Beijing where cleaning staff turned off lights and the robot continued working. The next day, they discovered it had completed the lobby cleaning overnight.
Office building cafeterias are also complex with frequently moving tables and chairs, and greasy floors. M360's non-repetitive scanning mode can clearly identify small obstacles, ensuring robots don't miss any corners.
Factory Environment
Factory environments put radar stability to the ultimate test. When multiple robots work together, M360 can perceive dynamic environments in real-time. During our tests, M360 could accurately identify operating AMRs and AGVs, automatically avoiding them. In such environments, M360's IP67 protection rating is particularly important - dust and oil don't affect its operation.
Summary and Recommendations
Frankly speaking, cleaning robots place more demanding requirements on LiDAR than other scenarios. They need strong near-field detection capabilities, long-term stable operation, and can't consume too much power.
From our six months of experience, M360's 5cm blind zone and 4.5W power consumption are the most impressive points. For cleaning scenarios, near-field detection capability is indeed more important than long-range detection. After all, cleaning robots mainly work indoors and rarely need to detect obstacles 50 meters away.
If your project is mainly in indoor environments, especially scenarios that need to handle shelf gaps and dense obstacles, I strongly recommend considering M360. Although it's heavier than MID-360, the improvements in stability and environmental adaptability are worth it.
To be honest, a robot's navigation performance is like a person's visual system - being able to see clearly and far is truly essential for good cleaning work. M360 gives cleaning robots a reliable pair of good eyes, so we no longer have to worry about robots getting lost in complex environments.