Revolutionizing Security: The Rise and Future of Security Robots
The idea of advanced security robots—intelligent machines designed to safeguard life and property, even at the risk of their own destruction—has, in popular culture, long captivated our imagination.
Growing up, I was particularly intrigued by Lost in Space, largely due to the compelling relationship between Will Robinson and "Robot" (designation B9). Robot exhibited a mix of protectiveness, loyalty, and mentorship towards young Will, which fostered a unique emotional bond between them.
This theme recurs in other iconic robots like Sonny from I, Robot; Baymax from Big Hero 6; the Iron Giant from the eponymous film; and the reprogrammed T-800 in Terminator 2. These fictional characters all share key traits: a protective instinct, loyalty, emotional connectivity with humans, and both sacrificial and heroic tendencies.
Are these qualities realistically applicable—or desirable—in security robots currently available or soon to enter the market? Furthermore, how do these qualities align with the tasks we expect these robots to perform? Would robots with artificial intelligence (AI) enabled, human-like behavior be more culturally and socially accepted in the workplace, and would the form factor—humanoid versus an R2D2-style—make any difference?
What exactly do we expect from today's security robots, and what might the future hold for them—and us?
Development of Sought-After Features
Security robots are distinct from the many other purpose-built robots in the workplace today. Robotics design for a security robot needs to factor in unique requirements like environment, intelligent and autonomous navigation in both indoor and outdoor uses, human interaction and communication requirements, sensory and perceptual capabilities, data security and privacy issues, and operational autonomy and decision-making.
Moreover, the AI model integrated into a security robot is crucial to its ability to perform the tasks we require with the level of autonomy and accuracy we expect. Today’s robots have very limited human interaction abilities but based on our organization’s Voice of the Customer initiatives, we’ve discovered that many end-users expect a high degree of human interaction, autonomous behavior, and decision-making performance.
Humanoid, non-humanoid, and aerial drones all fall within the spectrum of security robots. Each form factor is experiencing substantial development in the various technologies that enable them to meet our expectations while expanding their potential uses. Notably, builtin.com has identified 22 separate humanoid robots in various stages of development and appearance that can adopt a security application. This style of robots appears to be the future of security robotics.
Typical requirements for security robots include:
Autonomous Navigation: Since most security robots are not fixed in place, they must be equipped with one or more sophisticated navigation systems that may include camera vision, LiDAR, GPS, or SLAM (Simultaneous Localization and Mapping). These systems allow robots to autonomously move and patrol predefined areas while avoiding obstacles and hazards.
Audio and Visual Sensors: To identify and respond to sounds and visual cues, security robots need advanced sensors and AI capable of recognizing faces and interpreting body language. These capabilities enhance their ability to detect potential threats and interact with individuals effectively.
Human-Like Interaction: Humanoid security robots should be able to interact with people naturally and intuitively. This involves advanced speech recognition, natural language processing, and expressive communication abilities, including facial expressions and gestures.
Some of these requirements, such as recognizing and interpreting body language and its associated emotions, are still in the early stages of development. Progress is being made, driven partly by applications beyond physical security.
Likewise, several robotics companies are experimenting with facial modeling. They are designing robots to look as human as possible, including incorporating facial expressions, to enhance natural interactions with them. This endeavor is not without controversy. Some people feel uncomfortable with human-like machines, while others embrace the concept.
The progression of security robots from relatively simple machines to sophisticated, intelligent systems hinge on the development and integration of advanced AI models, notably large language model (LLMs). Currently, security robots are limited to basic tasks, such as patrolling predefined routes and reporting anomalies, with some robot interdictive behavior becoming possible. With the advent of LLMs and convolutional neural networks (CNNs), security robots are poised to become even more capable. These advanced AI technologies will allow robots to autonomously navigate increasingly complex environments, analyze vast amounts of data in real time, and make informed decisions independently.
By leveraging machine learning, computer vision, and natural language processing, robots will be capable of recognizing faces, interpreting body language, and interacting with humans in a more natural and intuitive manner. This progression will mark a significant leap from their earlier capabilities, setting the stage for a new era in security robotics where full autonomy and intelligence are at the forefront. This advancement not only enhances their effectiveness in detecting and mitigating threats but also allows them to adapt to dynamic situations, making them indispensable assets in modern security operations.
As AI technology continues to advance, we can expect security robots to become even more autonomous, intelligent, and integral to comprehensive security strategies.
Progress Toward System Integration and Overall Adoption
As security robots continue to advance in their specialized technologies, it will be crucial that they seamlessly integrate into broader security systems such as video surveillance, access control, and intrusion detection to maximize their functionality and efficiency.
This integration ensures that robots are not standalone units but part of a cohesive security strategy, enhancing their ability to communicate and coordinate with other measures autonomously at a systems level.
Additionally, the operational incorporation of these robots into a company’s security program is essential. Adopting security robots requires careful planning and organizational preparation, not only to leverage their current and future capabilities fully but also to address any workforce concerns. Since security robots represent a relatively new addition to the workplace environment, their presence can be unsettling for some employees. It is important to manage this transition thoughtfully, ensuring that all staff members are informed and comfortable with the robots’ roles in their daily work environment. Effective communication, training, and gradual integration can help build familiarity and trust between employees and robotic security systems, ensuring a smooth transition and optimal operational efficacy.
For example, one California-based aerospace company engaged its workforce by hosting socializing meetings with robots and employees. Employees were also asked to give each robot a unique name. The robots were quickly accepted into the company’s culture and the program is expected to grow to other sites as a standard security element.
Evolution and Impact of Large Language Models
The integration of LLMs into security robots heralds a significant evolution in how these machines interact within their operational environments. LLMs, with their increasingly sophisticated language processing and generation capabilities, offer the potential for security robots to develop unique personalities tailored to the cultural contexts of the workplaces they operate in. This customization could make robots more relatable and approachable, facilitating smoother social integration with human coworkers.
For instance, a security robot in a corporate office might adopt a formal tone and professional demeanor, while one in a creative industry setting could exhibit a more casual and friendly personality. Perhaps a robot located in Toronto cheers and announces when the Maple Leafs score against the Chicago Blackhawks!
Beyond personalization, LLMs could enhance security robots with several forward-looking capabilities. For example, they could enable advanced situational understanding, allowing robots to interpret nuances in human speech and behavior. This would allow robots to detect stress or anxiety in voices and body language, which might indicate potential security threats. This AI capability already exists in other technologies, including Amazon’s Alexa and IBM’s Watson.
Additionally, LLMs could improve decision-making processes by providing robots with the ability to analyze complex scenarios and generate multiple action strategies, assessing the implications of each in real time.
Another promising capability is continuous learning and adaptation, where robots can learn from ongoing interactions and feedback, progressively refining their responses and capabilities to align better with the evolving dynamics of their environment. This continuous learning would not only improve the effectiveness of security robots but also ensure they remain up to date with the latest security practices and threat assessments.
For example, if a patrolling robot learns that boxes have started to be piled in front of an emergency exit, it can remove the boxes, remember that the incident occurred, check all other exits for piles of boxes, and notify a human of the trend to correct employees’ behavior—just like a human guard would.
Robots, AI, and other emerging technologies are dramatically transforming how leaders can identify, control, and manage security risks. Never in our industry have rapidly evolving technologies offered such a profound impact on our roles and are poised to expand our human talent and resources. AI is central to this transformative wave, accompanied by advanced robotic developments and shifting cultural perceptions about this new and exciting technology.
The future is already upon us, and it promises to fundamentally alter how we protect enterprises, bringing profound and positive changes to our security strategies. The growing openness of the industry and end-users to emerging technologies, robotics, and AI-enabled solutions is a clear indicator of what’s to come—a seismic repositioning of technology within our enterprise-level security strategies, helping to leverage our human talent and resources more effectively while unlocking unheard-of potential for the future.
William Plante is the director, integrated solutions risk group, at Everon Solutions and a member of the ASIS International Emerging Technology Community Steering Committee.
© William Plante
