Sensible software program replaces costly sensors for glass wall detection with 96% accuracy

A analysis workforce has developed autonomous driving software program that permits cheap sensors to detect clear obstacles similar to glass partitions, offering an alternative choice to high-performance sensors. This expertise can be utilized in present robots, negating the necessity for added tools whereas making certain detection efficiency that is the same as that provided by costly typical tools.

The paper is revealed within the journal IEEE Transactions on Instrumentation and Measurement. The workforce was led by Professor Kyungjoon Park on the Division of Electrical Engineering and Laptop Science, Daegu Gyeongbuk Institute of Science & Know-how.

Autonomous driving robots sometimes use LiDAR sensors to detect their environment and navigate. Functioning as “laser eyes,” costly LiDAR sensors decide distance and construction by projecting gentle and measuring reflection time.

Cheap LiDAR sensors can’t detect clear objects similar to these fabricated from glass; they could mistake them for empty house, doubtlessly leading to a collision. Excessive-resolution ultrasonic LiDAR sensors or cameras wouldn’t have this limitation, however their use will increase system complexity and raises prices by a whole bunch of hundreds to hundreds of thousands of gained.

To offer an alternate, a DGIST analysis workforce led by Professor Kyungjoon Park developed probabilistic incremental navigation-based mapping (PINMAP), an algorithm that approaches problem-solving through software program, not {hardware}. PINMAP accumulates uncommon level knowledge that cheap LiDAR sensors can detect solely sporadically. Utilizing these knowledge, PINMAP probabilistically calculates the chance of the presence of glass partitions over time.

The PINMAP algorithm is predicated on Cartographer (map charting) and Nav2 (navigation), that are open-source instruments which are extensively used within the ROS 2 ecosystem. PINMAP has the benefit of straightforward applicability whereas eliminating the necessity to change the present system construction.

As an alternative of upgrading the sensors at a excessive price, the algorithm alters the best way the present sensors deal with knowledge; that’s, it makes use of software program to enhance the detection efficiency of cheap LiDAR sensors.

In a real-world experiment performed at DGIST, PINMAP detected glass partitions with 96.77% accuracy, which is effectively above the practically 0% detection price of the normal strategy utilizing the identical cheap LiDAR sensors (Cartographer-SLAM). The software program distinction that PINMAP gives demonstrated an amazing efficiency increase.

Professor Park stated, “PINMAP flips the traditional knowledge that {hardware} efficiency equals system efficiency and proposes a brand new commonplace whereby software program can enhance sensor capabilities. This examine reveals that making certain steady autonomous driving is feasible with out counting on high-performance tools.”

The algorithm the analysis workforce developed gives a considerable financial benefit as a result of it achieves detection efficiency akin to that of pricey LiDAR sensors at lower than one-tenth of the fee. This expertise is anticipated to cut back collisions between autonomous driving robots and glass or clear acrylic partitions in indoor areas similar to hospitals, airports, purchasing malls, and warehouses, thus contributing to the large-scale deployment of service robots.