In the past six months, I've helped deploy dozens of mobile robot projects for friends, and the pitfalls I've encountered are more than the code I've written. The most frustrating part is not the integration and debugging, but rather when the robot suddenly "goes blind."
For example, we provided cleaning robots for a catering chain, and initially, the laser radar cleaned the restaurant environment well. Later, another robot was installed next door, and the problem arose—the signals of the two LiDARs interfered with each other, the point cloud data was full of noise, and the robot moved around like a person who had been drinking. Worse still, the glass door at the kitchen entrance could not be penetrated by the single echo radar, and it collided with the glass several times.
Later, we switched to the M360 dual echo version, and things were completely different. Today, let's talk about this technical detail.
The "Vision Problem" of LiDAR
Traditional LiDAR has several obvious shortcomings in complex environments:
Signal interference problem
Placing two LiDARs in the same space is like two people speaking at the same time and not being able to hear each other. A real case we encountered: a cleaning robot installed on the first floor and second floor of a restaurant interfered with each other as soon as they were turned on, resulting in chaotic point cloud data and even failing to avoid obstacles.
It's also a hassle for single echo radar to encounter glass, such as the glass doors of fast-food restaurants, display cabinets, etc. The radar will avoid the glass as an obstacle, but cannot penetrate it, causing the robot to never be able to avoid these areas.
Environmental Light Influence
Indoors are fine, but outdoors can be troublesome.
M360's solution
Dual echo penetration technology
The M360 can receive both forward and backward echoes simultaneously, which means it can "see" the front of obstacles and "penetrate" through raindrops, glass, and see the situation behind.
Test data: On rainy days, the M360 can still stably detect pedestrians 3 meters away, with a penetration rate of 95%. In comparison, a single echo radar may not even guarantee a distance of 1 meter in this environment, and the point cloud data is full of noise.
IP67 protection rating
Waterproof and dustproof are not just talk. We have tested it in the factory; once, a cleaning robot accidentally hit the edge of a sink and got wet, but it was retrieved and continued to be used without any issues. This 6.5cm small sphere can indeed withstand certain physical and environmental impacts.
5cm ultra-near blind zone
This parameter is extremely critical in real-world scenarios. Many small obstacles, such as gaps between warehouse shelves, the bottom of factory equipment, and the legs of dining tables, are only 3-5cm from the ground. The 10cm blind zone of traditional radars is simply unable to detect them. The 5cm blind zone of the M360 means that the robot can "see" all obstacles in complex spaces.
Actual application comparison
In the catering scenario, the advantages of the M360 are particularly obvious:
Multi-robot collaboration
With five robots working simultaneously in the same space, the M360's dual echo mode can automatically filter out interference from other radars, and the point cloud data remains clean and accurate. This is especially important for unified deployment in chain stores.
Adaptability to complex environments
The steam in the kitchen, the lighting changes in restaurants, the reflection of metal shelves in supermarkets, these environmental factors are a challenge for ordinary radars. The M360 performs stably in various environments and will not lose its target due to environmental changes.
Cost-effectiveness
Although the M360 is not cheap, considering the reduced failure rate and maintenance costs, the total cost of ownership is actually lower. One customer calculated that after using the M360, the robot's failure rate dropped by 60%, and the savings on maintenance costs in a year could cover the cost of several LiDARs.
Experience Summary
Over the years of working on robot projects, the biggest feeling is: don't believe in laboratory data, look at the actual scenario. Sometimes the radar with the longest ranging distance on the parameter table may not be as good as those that seem ordinary but have strong interference resistance in actual application.
The greatest value of the M360 is not its impressive parameters, but its ability to work stably in various chaotic environments. For commercial robots, stability and reliability are much more important than being "parameter-leading."
If you are also considering the selection of robot LiDAR, it is recommended to go to the actual site for testing to see how it performs in the real environment. After all, good-looking parameters are not as good as stable and practical performance.